{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.24,
  "eval_steps": 500,
  "global_step": 450,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "completion_length": 496.2109546661377,
      "epoch": 0.0010666666666666667,
      "grad_norm": 0.11954631782793526,
      "kl": 0.0,
      "learning_rate": 7.142857142857142e-08,
      "loss": -0.0,
      "reward": 0.3255208423361182,
      "reward_std": 0.41301608085632324,
      "rewards/equation_reward_func": 0.0729166679084301,
      "rewards/format_reward_func": 0.2526041716337204,
      "step": 2
    },
    {
      "completion_length": 502.083345413208,
      "epoch": 0.0021333333333333334,
      "grad_norm": 0.13612224468593256,
      "kl": 0.0004134178161621094,
      "learning_rate": 1.4285714285714285e-07,
      "loss": 0.0,
      "reward": 0.32552084419876337,
      "reward_std": 0.46909805946052074,
      "rewards/equation_reward_func": 0.07031250256113708,
      "rewards/format_reward_func": 0.2552083390764892,
      "step": 4
    },
    {
      "completion_length": 478.82554054260254,
      "epoch": 0.0032,
      "grad_norm": 0.14255152767670767,
      "kl": 0.0003870725631713867,
      "learning_rate": 2.1428571428571426e-07,
      "loss": 0.0,
      "reward": 0.32291667349636555,
      "reward_std": 0.5039449613541365,
      "rewards/equation_reward_func": 0.06770833465270698,
      "rewards/format_reward_func": 0.25520834140479565,
      "step": 6
    },
    {
      "completion_length": 488.739595413208,
      "epoch": 0.004266666666666667,
      "grad_norm": 0.13619929714048676,
      "kl": 0.00038945674896240234,
      "learning_rate": 2.857142857142857e-07,
      "loss": 0.0,
      "reward": 0.33333334047347307,
      "reward_std": 0.4558701850473881,
      "rewards/equation_reward_func": 0.054687500931322575,
      "rewards/format_reward_func": 0.2786458423361182,
      "step": 8
    },
    {
      "completion_length": 489.1224060058594,
      "epoch": 0.005333333333333333,
      "grad_norm": 0.13326587315385535,
      "kl": 0.0003973245620727539,
      "learning_rate": 3.5714285714285716e-07,
      "loss": 0.0,
      "reward": 0.37760417629033327,
      "reward_std": 0.4923101980239153,
      "rewards/equation_reward_func": 0.03645833395421505,
      "rewards/format_reward_func": 0.3411458460614085,
      "step": 10
    },
    {
      "completion_length": 493.63543128967285,
      "epoch": 0.0064,
      "grad_norm": 0.14488433794501374,
      "kl": 0.0005027055740356445,
      "learning_rate": 4.285714285714285e-07,
      "loss": 0.0,
      "reward": 0.3333333423361182,
      "reward_std": 0.4829239808022976,
      "rewards/equation_reward_func": 0.0416666679084301,
      "rewards/format_reward_func": 0.2916666753590107,
      "step": 12
    },
    {
      "completion_length": 495.4921989440918,
      "epoch": 0.007466666666666667,
      "grad_norm": 0.13465334279118427,
      "kl": 0.0008780956268310547,
      "learning_rate": 5e-07,
      "loss": 0.0,
      "reward": 0.4635416828095913,
      "reward_std": 0.533896965906024,
      "rewards/equation_reward_func": 0.05989583441987634,
      "rewards/format_reward_func": 0.4036458469927311,
      "step": 14
    },
    {
      "completion_length": 451.22657203674316,
      "epoch": 0.008533333333333334,
      "grad_norm": 0.15101250146425993,
      "kl": 0.0014376640319824219,
      "learning_rate": 4.999740409224932e-07,
      "loss": 0.0,
      "reward": 0.5651041828095913,
      "reward_std": 0.5559330955147743,
      "rewards/equation_reward_func": 0.06510416767559946,
      "rewards/format_reward_func": 0.500000013038516,
      "step": 16
    },
    {
      "completion_length": 465.8593864440918,
      "epoch": 0.0096,
      "grad_norm": 0.10999101009073732,
      "kl": 0.0033292770385742188,
      "learning_rate": 4.998961690809627e-07,
      "loss": 0.0,
      "reward": 0.661458345130086,
      "reward_std": 0.5498605705797672,
      "rewards/equation_reward_func": 0.07552083651535213,
      "rewards/format_reward_func": 0.585937513038516,
      "step": 18
    },
    {
      "completion_length": 484.29168128967285,
      "epoch": 0.010666666666666666,
      "grad_norm": 0.10614223340608818,
      "kl": 0.005473136901855469,
      "learning_rate": 4.997664006472578e-07,
      "loss": 0.0,
      "reward": 0.7630208544433117,
      "reward_std": 0.4911875668913126,
      "rewards/equation_reward_func": 0.054687501629814506,
      "rewards/format_reward_func": 0.7083333507180214,
      "step": 20
    },
    {
      "completion_length": 459.4166851043701,
      "epoch": 0.011733333333333333,
      "grad_norm": 0.10241538861175935,
      "kl": 0.0071125030517578125,
      "learning_rate": 4.995847625707292e-07,
      "loss": 0.0,
      "reward": 0.8463541828095913,
      "reward_std": 0.4817237760871649,
      "rewards/equation_reward_func": 0.06770833488553762,
      "rewards/format_reward_func": 0.778645858168602,
      "step": 22
    },
    {
      "completion_length": 460.716157913208,
      "epoch": 0.0128,
      "grad_norm": 0.08441145858520885,
      "kl": 0.0069980621337890625,
      "learning_rate": 4.993512925726318e-07,
      "loss": 0.0,
      "reward": 0.8906250298023224,
      "reward_std": 0.4445338100194931,
      "rewards/equation_reward_func": 0.07812500256113708,
      "rewards/format_reward_func": 0.8125000186264515,
      "step": 24
    },
    {
      "completion_length": 483.37761878967285,
      "epoch": 0.013866666666666666,
      "grad_norm": 0.09439729466177939,
      "kl": 0.007844924926757812,
      "learning_rate": 4.990660391382923e-07,
      "loss": 0.0,
      "reward": 0.8984375149011612,
      "reward_std": 0.4334138287231326,
      "rewards/equation_reward_func": 0.06770833535119891,
      "rewards/format_reward_func": 0.8307291865348816,
      "step": 26
    },
    {
      "completion_length": 487.11720085144043,
      "epoch": 0.014933333333333333,
      "grad_norm": 0.07829204100262112,
      "kl": 0.008859634399414062,
      "learning_rate": 4.987290615070384e-07,
      "loss": 0.0,
      "reward": 0.9244792051613331,
      "reward_std": 0.36052799224853516,
      "rewards/equation_reward_func": 0.07291666860692203,
      "rewards/format_reward_func": 0.8515625186264515,
      "step": 28
    },
    {
      "completion_length": 485.40625953674316,
      "epoch": 0.016,
      "grad_norm": 0.07302798822426318,
      "kl": 0.0127105712890625,
      "learning_rate": 4.983404296598978e-07,
      "loss": 0.0,
      "reward": 0.9479167014360428,
      "reward_std": 0.33515751641243696,
      "rewards/equation_reward_func": 0.06510416837409139,
      "rewards/format_reward_func": 0.8828125111758709,
      "step": 30
    },
    {
      "completion_length": 450.005220413208,
      "epoch": 0.017066666666666667,
      "grad_norm": 0.09195588444312232,
      "kl": 0.013427734375,
      "learning_rate": 4.979002243050646e-07,
      "loss": 0.0,
      "reward": 1.0833333656191826,
      "reward_std": 0.39109824877232313,
      "rewards/equation_reward_func": 0.15104167233221233,
      "rewards/format_reward_func": 0.9322916828095913,
      "step": 32
    },
    {
      "completion_length": 450.716157913208,
      "epoch": 0.018133333333333335,
      "grad_norm": 0.07847022887495707,
      "kl": 0.01331329345703125,
      "learning_rate": 4.974085368611381e-07,
      "loss": 0.0,
      "reward": 1.057291690260172,
      "reward_std": 0.32324595795944333,
      "rewards/equation_reward_func": 0.12239583651535213,
      "rewards/format_reward_func": 0.9348958544433117,
      "step": 34
    },
    {
      "completion_length": 460.44792556762695,
      "epoch": 0.0192,
      "grad_norm": 0.07189632688173346,
      "kl": 0.015232086181640625,
      "learning_rate": 4.968654694381379e-07,
      "loss": 0.0,
      "reward": 1.0520833805203438,
      "reward_std": 0.3206865997053683,
      "rewards/equation_reward_func": 0.1145833374466747,
      "rewards/format_reward_func": 0.9375000149011612,
      "step": 36
    },
    {
      "completion_length": 433.51563453674316,
      "epoch": 0.020266666666666665,
      "grad_norm": 0.08532850510429227,
      "kl": 0.01625823974609375,
      "learning_rate": 4.962711348162987e-07,
      "loss": 0.0,
      "reward": 1.0833333656191826,
      "reward_std": 0.32565778447315097,
      "rewards/equation_reward_func": 0.1328125037252903,
      "rewards/format_reward_func": 0.9505208507180214,
      "step": 38
    },
    {
      "completion_length": 440.76042556762695,
      "epoch": 0.021333333333333333,
      "grad_norm": 0.0822050428393239,
      "kl": 0.016002655029296875,
      "learning_rate": 4.956256564226487e-07,
      "loss": 0.0,
      "reward": 1.0520833618938923,
      "reward_std": 0.3361439718864858,
      "rewards/equation_reward_func": 0.11979167070239782,
      "rewards/format_reward_func": 0.9322916977107525,
      "step": 40
    },
    {
      "completion_length": 416.8619899749756,
      "epoch": 0.0224,
      "grad_norm": 0.06432799392042395,
      "kl": 0.0177764892578125,
      "learning_rate": 4.949291683053768e-07,
      "loss": 0.0,
      "reward": 1.0729166939854622,
      "reward_std": 0.2632621508091688,
      "rewards/equation_reward_func": 0.1119791695382446,
      "rewards/format_reward_func": 0.9609375260770321,
      "step": 42
    },
    {
      "completion_length": 407.489595413208,
      "epoch": 0.023466666666666667,
      "grad_norm": 0.0816513301308808,
      "kl": 0.019321441650390625,
      "learning_rate": 4.941818151059955e-07,
      "loss": 0.0,
      "reward": 1.0859375298023224,
      "reward_std": 0.2334962603636086,
      "rewards/equation_reward_func": 0.1067708374466747,
      "rewards/format_reward_func": 0.9791666865348816,
      "step": 44
    },
    {
      "completion_length": 437.7265739440918,
      "epoch": 0.024533333333333334,
      "grad_norm": 0.07098712073151407,
      "kl": 0.020320892333984375,
      "learning_rate": 4.933837520293017e-07,
      "loss": 0.0,
      "reward": 1.0494791977107525,
      "reward_std": 0.24554855143651366,
      "rewards/equation_reward_func": 0.08593750093132257,
      "rewards/format_reward_func": 0.963541679084301,
      "step": 46
    },
    {
      "completion_length": 409.68750953674316,
      "epoch": 0.0256,
      "grad_norm": 0.08052848471910203,
      "kl": 0.01833343505859375,
      "learning_rate": 4.925351448111454e-07,
      "loss": 0.0,
      "reward": 1.1171875335276127,
      "reward_std": 0.3026689598336816,
      "rewards/equation_reward_func": 0.14583333558402956,
      "rewards/format_reward_func": 0.971354179084301,
      "step": 48
    },
    {
      "completion_length": 408.2890739440918,
      "epoch": 0.02666666666666667,
      "grad_norm": 0.06883175239258556,
      "kl": 0.0229949951171875,
      "learning_rate": 4.91636169684011e-07,
      "loss": 0.0,
      "reward": 1.0859375409781933,
      "reward_std": 0.2610642076469958,
      "rewards/equation_reward_func": 0.12500000349245965,
      "rewards/format_reward_func": 0.9609375149011612,
      "step": 50
    },
    {
      "completion_length": 406.5026149749756,
      "epoch": 0.027733333333333332,
      "grad_norm": 0.08432747124886808,
      "kl": 0.02112579345703125,
      "learning_rate": 4.906870133404186e-07,
      "loss": 0.0,
      "reward": 1.1276042088866234,
      "reward_std": 0.30215284787118435,
      "rewards/equation_reward_func": 0.1588541716337204,
      "rewards/format_reward_func": 0.9687500186264515,
      "step": 52
    },
    {
      "completion_length": 414.50782203674316,
      "epoch": 0.0288,
      "grad_norm": 0.08259119762071515,
      "kl": 0.02127838134765625,
      "learning_rate": 4.896878728941531e-07,
      "loss": 0.0,
      "reward": 1.0755208618938923,
      "reward_std": 0.27385374438017607,
      "rewards/equation_reward_func": 0.1119791695382446,
      "rewards/format_reward_func": 0.9635416865348816,
      "step": 54
    },
    {
      "completion_length": 403.9401149749756,
      "epoch": 0.029866666666666666,
      "grad_norm": 0.06375644666852129,
      "kl": 0.0231781005859375,
      "learning_rate": 4.886389558393284e-07,
      "loss": 0.0,
      "reward": 1.093750026077032,
      "reward_std": 0.23606119910255075,
      "rewards/equation_reward_func": 0.12239583674818277,
      "rewards/format_reward_func": 0.971354179084301,
      "step": 56
    },
    {
      "completion_length": 399.2760524749756,
      "epoch": 0.030933333333333334,
      "grad_norm": 0.08378363145506801,
      "kl": 0.02516937255859375,
      "learning_rate": 4.875404800072976e-07,
      "loss": 0.0,
      "reward": 1.1015625298023224,
      "reward_std": 0.25781850097700953,
      "rewards/equation_reward_func": 0.1276041683740914,
      "rewards/format_reward_func": 0.9739583432674408,
      "step": 58
    },
    {
      "completion_length": 385.61719703674316,
      "epoch": 0.032,
      "grad_norm": 0.07405086579557206,
      "kl": 0.02643585205078125,
      "learning_rate": 4.86392673521415e-07,
      "loss": 0.0,
      "reward": 1.1354167088866234,
      "reward_std": 0.2644475535489619,
      "rewards/equation_reward_func": 0.1562500053551048,
      "rewards/format_reward_func": 0.9791666753590107,
      "step": 60
    },
    {
      "completion_length": 388.5390682220459,
      "epoch": 0.03306666666666667,
      "grad_norm": 0.07936857349188958,
      "kl": 0.0322113037109375,
      "learning_rate": 4.851957747496606e-07,
      "loss": 0.0,
      "reward": 1.1093750409781933,
      "reward_std": 0.24481901014223695,
      "rewards/equation_reward_func": 0.13020833767950535,
      "rewards/format_reward_func": 0.9791666828095913,
      "step": 62
    },
    {
      "completion_length": 374.6614685058594,
      "epoch": 0.034133333333333335,
      "grad_norm": 0.08002757038050164,
      "kl": 0.0284271240234375,
      "learning_rate": 4.839500322551386e-07,
      "loss": 0.0,
      "reward": 1.1302083693444729,
      "reward_std": 0.2239288128912449,
      "rewards/equation_reward_func": 0.14322917046956718,
      "rewards/format_reward_func": 0.9869791753590107,
      "step": 64
    },
    {
      "completion_length": 374.56250953674316,
      "epoch": 0.0352,
      "grad_norm": 0.08882377481939042,
      "kl": 0.03044891357421875,
      "learning_rate": 4.826557047444563e-07,
      "loss": 0.0,
      "reward": 1.1250000298023224,
      "reward_std": 0.25083459448069334,
      "rewards/equation_reward_func": 0.14322917070239782,
      "rewards/format_reward_func": 0.9817708395421505,
      "step": 66
    },
    {
      "completion_length": 360.7890748977661,
      "epoch": 0.03626666666666667,
      "grad_norm": 0.09104736623524047,
      "kl": 0.03575897216796875,
      "learning_rate": 4.813130610139993e-07,
      "loss": 0.0,
      "reward": 1.1588542014360428,
      "reward_std": 0.24613964278250933,
      "rewards/equation_reward_func": 0.16927083674818277,
      "rewards/format_reward_func": 0.9895833432674408,
      "step": 68
    },
    {
      "completion_length": 363.380220413208,
      "epoch": 0.037333333333333336,
      "grad_norm": 0.08230130539588183,
      "kl": 0.03208160400390625,
      "learning_rate": 4.799223798941089e-07,
      "loss": 0.0,
      "reward": 1.1380208805203438,
      "reward_std": 0.26733159739524126,
      "rewards/equation_reward_func": 0.16406250488944352,
      "rewards/format_reward_func": 0.9739583507180214,
      "step": 70
    },
    {
      "completion_length": 344.23959255218506,
      "epoch": 0.0384,
      "grad_norm": 0.10598274417848667,
      "kl": 0.0363922119140625,
      "learning_rate": 4.78483950191177e-07,
      "loss": 0.0,
      "reward": 1.1901042014360428,
      "reward_std": 0.2702939258888364,
      "rewards/equation_reward_func": 0.19531250488944352,
      "rewards/format_reward_func": 0.9947916716337204,
      "step": 72
    },
    {
      "completion_length": 355.94271659851074,
      "epoch": 0.039466666666666664,
      "grad_norm": 0.0868727954216351,
      "kl": 0.0372161865234375,
      "learning_rate": 4.769980706276687e-07,
      "loss": 0.0,
      "reward": 1.1276042088866234,
      "reward_std": 0.24663167539983988,
      "rewards/equation_reward_func": 0.14583334070630372,
      "rewards/format_reward_func": 0.9817708432674408,
      "step": 74
    },
    {
      "completion_length": 377.73959732055664,
      "epoch": 0.04053333333333333,
      "grad_norm": 0.07917788675218197,
      "kl": 0.0357513427734375,
      "learning_rate": 4.7546504978008595e-07,
      "loss": 0.0,
      "reward": 1.1432292088866234,
      "reward_std": 0.24032680923119187,
      "rewards/equation_reward_func": 0.1536458374466747,
      "rewards/format_reward_func": 0.9895833432674408,
      "step": 76
    },
    {
      "completion_length": 360.8307399749756,
      "epoch": 0.0416,
      "grad_norm": 0.09413557754270317,
      "kl": 0.038055419921875,
      "learning_rate": 4.738852060148848e-07,
      "loss": 0.0,
      "reward": 1.1796875521540642,
      "reward_std": 0.3384167607873678,
      "rewards/equation_reward_func": 0.20833333837799728,
      "rewards/format_reward_func": 0.971354179084301,
      "step": 78
    },
    {
      "completion_length": 361.27084159851074,
      "epoch": 0.042666666666666665,
      "grad_norm": 0.09743025798738646,
      "kl": 0.04107666015625,
      "learning_rate": 4.722588674223593e-07,
      "loss": 0.0,
      "reward": 1.2343750447034836,
      "reward_std": 0.3094023009762168,
      "rewards/equation_reward_func": 0.24739584070630372,
      "rewards/format_reward_func": 0.986979179084301,
      "step": 80
    },
    {
      "completion_length": 368.3567810058594,
      "epoch": 0.04373333333333333,
      "grad_norm": 0.0716687709123018,
      "kl": 0.04241943359375,
      "learning_rate": 4.70586371748506e-07,
      "loss": 0.0,
      "reward": 1.119791705161333,
      "reward_std": 0.2431359770707786,
      "rewards/equation_reward_func": 0.13541667046956718,
      "rewards/format_reward_func": 0.9843750074505806,
      "step": 82
    },
    {
      "completion_length": 373.70834159851074,
      "epoch": 0.0448,
      "grad_norm": 0.08456888547469173,
      "kl": 0.043853759765625,
      "learning_rate": 4.6886806632488363e-07,
      "loss": 0.0,
      "reward": 1.2239583805203438,
      "reward_std": 0.32931508449837565,
      "rewards/equation_reward_func": 0.2500000074505806,
      "rewards/format_reward_func": 0.9739583507180214,
      "step": 84
    },
    {
      "completion_length": 363.72136306762695,
      "epoch": 0.04586666666666667,
      "grad_norm": 0.09097792518227367,
      "kl": 0.0495147705078125,
      "learning_rate": 4.6710430799648143e-07,
      "loss": 0.0,
      "reward": 1.2343750447034836,
      "reward_std": 0.3150203409604728,
      "rewards/equation_reward_func": 0.25260417466051877,
      "rewards/format_reward_func": 0.9817708469927311,
      "step": 86
    },
    {
      "completion_length": 368.83594703674316,
      "epoch": 0.046933333333333334,
      "grad_norm": 0.08075857490244749,
      "kl": 0.04754638671875,
      "learning_rate": 4.652954630476127e-07,
      "loss": 0.0,
      "reward": 1.226562537252903,
      "reward_std": 0.33071469189599156,
      "rewards/equation_reward_func": 0.247395837912336,
      "rewards/format_reward_func": 0.9791666828095913,
      "step": 88
    },
    {
      "completion_length": 383.08073806762695,
      "epoch": 0.048,
      "grad_norm": 0.07265214510146589,
      "kl": 0.0546112060546875,
      "learning_rate": 4.6344190712584713e-07,
      "loss": 0.0001,
      "reward": 1.2057291977107525,
      "reward_std": 0.2948309718631208,
      "rewards/equation_reward_func": 0.231770841171965,
      "rewards/format_reward_func": 0.9739583432674408,
      "step": 90
    },
    {
      "completion_length": 383.3177185058594,
      "epoch": 0.04906666666666667,
      "grad_norm": 0.1051600515841224,
      "kl": 0.052886962890625,
      "learning_rate": 4.615440251639995e-07,
      "loss": 0.0001,
      "reward": 1.229166716337204,
      "reward_std": 0.33057287614792585,
      "rewards/equation_reward_func": 0.24739584000781178,
      "rewards/format_reward_func": 0.9817708469927311,
      "step": 92
    },
    {
      "completion_length": 404.4583396911621,
      "epoch": 0.050133333333333335,
      "grad_norm": 0.0785283851923137,
      "kl": 0.0536956787109375,
      "learning_rate": 4.596022113001894e-07,
      "loss": 0.0001,
      "reward": 1.1718750298023224,
      "reward_std": 0.2629168755374849,
      "rewards/equation_reward_func": 0.19531250256113708,
      "rewards/format_reward_func": 0.9765625149011612,
      "step": 94
    },
    {
      "completion_length": 393.22136306762695,
      "epoch": 0.0512,
      "grad_norm": 0.08425171673805154,
      "kl": 0.06390380859375,
      "learning_rate": 4.576168687959895e-07,
      "loss": 0.0001,
      "reward": 1.2109375447034836,
      "reward_std": 0.31305414671078324,
      "rewards/equation_reward_func": 0.23437500419095159,
      "rewards/format_reward_func": 0.9765625149011612,
      "step": 96
    },
    {
      "completion_length": 380.39584827423096,
      "epoch": 0.05226666666666667,
      "grad_norm": 0.0719108143516897,
      "kl": 0.0635528564453125,
      "learning_rate": 4.555884099526793e-07,
      "loss": 0.0001,
      "reward": 1.255208358168602,
      "reward_std": 0.2745982948690653,
      "rewards/equation_reward_func": 0.26822917140088975,
      "rewards/format_reward_func": 0.9869791753590107,
      "step": 98
    },
    {
      "completion_length": 382.8046989440918,
      "epoch": 0.05333333333333334,
      "grad_norm": 0.0824603010214514,
      "kl": 0.0673675537109375,
      "learning_rate": 4.5351725602562174e-07,
      "loss": 0.0001,
      "reward": 1.2447917014360428,
      "reward_std": 0.28847640473395586,
      "rewards/equation_reward_func": 0.26822917349636555,
      "rewards/format_reward_func": 0.9765625111758709,
      "step": 100
    },
    {
      "completion_length": 442.31771659851074,
      "epoch": 0.0544,
      "grad_norm": 0.07580652402214415,
      "kl": 0.063507080078125,
      "learning_rate": 4.514038371367791e-07,
      "loss": 0.0001,
      "reward": 1.213541705161333,
      "reward_std": 0.3326714150607586,
      "rewards/equation_reward_func": 0.25260417303070426,
      "rewards/format_reward_func": 0.9609375111758709,
      "step": 102
    },
    {
      "completion_length": 366.58855056762695,
      "epoch": 0.055466666666666664,
      "grad_norm": 0.08608427800244613,
      "kl": 0.0719146728515625,
      "learning_rate": 4.4924859218538936e-07,
      "loss": 0.0001,
      "reward": 1.2786458730697632,
      "reward_std": 0.3109776326455176,
      "rewards/equation_reward_func": 0.2994791774544865,
      "rewards/format_reward_func": 0.9791666828095913,
      "step": 104
    },
    {
      "completion_length": 397.6562623977661,
      "epoch": 0.05653333333333333,
      "grad_norm": 0.08077416683670491,
      "kl": 0.0741424560546875,
      "learning_rate": 4.470519687568185e-07,
      "loss": 0.0001,
      "reward": 1.236979190260172,
      "reward_std": 0.306100333109498,
      "rewards/equation_reward_func": 0.26822917466051877,
      "rewards/format_reward_func": 0.9687500223517418,
      "step": 106
    },
    {
      "completion_length": 392.052095413208,
      "epoch": 0.0576,
      "grad_norm": 0.11099671542231605,
      "kl": 0.072998046875,
      "learning_rate": 4.4481442302960923e-07,
      "loss": 0.0001,
      "reward": 1.2760416939854622,
      "reward_std": 0.3399705649353564,
      "rewards/equation_reward_func": 0.2994791765231639,
      "rewards/format_reward_func": 0.9765625149011612,
      "step": 108
    },
    {
      "completion_length": 370.3906316757202,
      "epoch": 0.058666666666666666,
      "grad_norm": 0.09213409066857482,
      "kl": 0.08636474609375,
      "learning_rate": 4.4253641968074505e-07,
      "loss": 0.0001,
      "reward": 1.3437500298023224,
      "reward_std": 0.3902808427810669,
      "rewards/equation_reward_func": 0.38281250931322575,
      "rewards/format_reward_func": 0.9609375149011612,
      "step": 110
    },
    {
      "completion_length": 420.138032913208,
      "epoch": 0.05973333333333333,
      "grad_norm": 0.07358301230035058,
      "kl": 0.081298828125,
      "learning_rate": 4.402184317891501e-07,
      "loss": 0.0001,
      "reward": 1.2005208805203438,
      "reward_std": 0.28474497282877564,
      "rewards/equation_reward_func": 0.23437500977888703,
      "rewards/format_reward_func": 0.9661458544433117,
      "step": 112
    },
    {
      "completion_length": 432.9974060058594,
      "epoch": 0.0608,
      "grad_norm": 0.069887196509218,
      "kl": 0.0760345458984375,
      "learning_rate": 4.37860940737443e-07,
      "loss": 0.0001,
      "reward": 1.1875000335276127,
      "reward_std": 0.2837864141911268,
      "rewards/equation_reward_func": 0.23697917396202683,
      "rewards/format_reward_func": 0.9505208507180214,
      "step": 114
    },
    {
      "completion_length": 422.9088649749756,
      "epoch": 0.06186666666666667,
      "grad_norm": 0.09261482901163447,
      "kl": 0.081817626953125,
      "learning_rate": 4.354644361119671e-07,
      "loss": 0.0001,
      "reward": 1.2031250298023224,
      "reward_std": 0.2831558440811932,
      "rewards/equation_reward_func": 0.25260417303070426,
      "rewards/format_reward_func": 0.9505208544433117,
      "step": 116
    },
    {
      "completion_length": 396.2109489440918,
      "epoch": 0.06293333333333333,
      "grad_norm": 0.084517772171227,
      "kl": 0.0916595458984375,
      "learning_rate": 4.3302941560111716e-07,
      "loss": 0.0001,
      "reward": 1.2864583618938923,
      "reward_std": 0.2684871470555663,
      "rewards/equation_reward_func": 0.31510417349636555,
      "rewards/format_reward_func": 0.9713541716337204,
      "step": 118
    },
    {
      "completion_length": 406.57813358306885,
      "epoch": 0.064,
      "grad_norm": 0.09081750104097751,
      "kl": 0.087799072265625,
      "learning_rate": 4.3055638489198236e-07,
      "loss": 0.0001,
      "reward": 1.2630208805203438,
      "reward_std": 0.3080101846717298,
      "rewards/equation_reward_func": 0.2890625100117177,
      "rewards/format_reward_func": 0.9739583507180214,
      "step": 120
    },
    {
      "completion_length": 355.86459255218506,
      "epoch": 0.06506666666666666,
      "grad_norm": 0.08971890990350641,
      "kl": 0.099761962890625,
      "learning_rate": 4.280458575653296e-07,
      "loss": 0.0001,
      "reward": 1.3619792014360428,
      "reward_std": 0.33915683487430215,
      "rewards/equation_reward_func": 0.38541667675599456,
      "rewards/format_reward_func": 0.9765625074505806,
      "step": 122
    },
    {
      "completion_length": 390.567720413208,
      "epoch": 0.06613333333333334,
      "grad_norm": 0.07864318983071089,
      "kl": 0.09515380859375,
      "learning_rate": 4.2549835498894665e-07,
      "loss": 0.0001,
      "reward": 1.338541705161333,
      "reward_std": 0.29398106783628464,
      "rewards/equation_reward_func": 0.3697916779201478,
      "rewards/format_reward_func": 0.9687500186264515,
      "step": 124
    },
    {
      "completion_length": 410.3906364440918,
      "epoch": 0.0672,
      "grad_norm": 0.08089125478754702,
      "kl": 0.12396240234375,
      "learning_rate": 4.229144062093679e-07,
      "loss": 0.0001,
      "reward": 1.3229167014360428,
      "reward_std": 0.2598429606296122,
      "rewards/equation_reward_func": 0.3411458423361182,
      "rewards/format_reward_func": 0.9817708432674408,
      "step": 126
    },
    {
      "completion_length": 403.84896659851074,
      "epoch": 0.06826666666666667,
      "grad_norm": 0.06893337797991321,
      "kl": 0.11248779296875,
      "learning_rate": 4.2029454784200675e-07,
      "loss": 0.0001,
      "reward": 1.3437500447034836,
      "reward_std": 0.29728816309943795,
      "rewards/equation_reward_func": 0.37239584303461015,
      "rewards/format_reward_func": 0.9713541865348816,
      "step": 128
    },
    {
      "completion_length": 424.90365409851074,
      "epoch": 0.06933333333333333,
      "grad_norm": 0.07036206011933169,
      "kl": 0.094207763671875,
      "learning_rate": 4.1763932395971433e-07,
      "loss": 0.0001,
      "reward": 1.2317708767950535,
      "reward_std": 0.271359003148973,
      "rewards/equation_reward_func": 0.2656250079162419,
      "rewards/format_reward_func": 0.9661458432674408,
      "step": 130
    },
    {
      "completion_length": 427.7578239440918,
      "epoch": 0.0704,
      "grad_norm": 0.08260904981702806,
      "kl": 0.097991943359375,
      "learning_rate": 4.1494928597979117e-07,
      "loss": 0.0001,
      "reward": 1.3203125409781933,
      "reward_std": 0.35679714987054467,
      "rewards/equation_reward_func": 0.36718751303851604,
      "rewards/format_reward_func": 0.9531250111758709,
      "step": 132
    },
    {
      "completion_length": 434.59896659851074,
      "epoch": 0.07146666666666666,
      "grad_norm": 0.07272128415662933,
      "kl": 0.099395751953125,
      "learning_rate": 4.122249925494726e-07,
      "loss": 0.0001,
      "reward": 1.2864583656191826,
      "reward_std": 0.2933720201253891,
      "rewards/equation_reward_func": 0.333333341171965,
      "rewards/format_reward_func": 0.9531250223517418,
      "step": 134
    },
    {
      "completion_length": 441.93230628967285,
      "epoch": 0.07253333333333334,
      "grad_norm": 0.05907301307740773,
      "kl": 0.1224365234375,
      "learning_rate": 4.094670094299131e-07,
      "loss": 0.0001,
      "reward": 1.299479205161333,
      "reward_std": 0.27806125255301595,
      "rewards/equation_reward_func": 0.3385416797827929,
      "rewards/format_reward_func": 0.9609375186264515,
      "step": 136
    },
    {
      "completion_length": 423.520845413208,
      "epoch": 0.0736,
      "grad_norm": 0.07954765389278069,
      "kl": 0.099212646484375,
      "learning_rate": 4.066759093786931e-07,
      "loss": 0.0001,
      "reward": 1.296875037252903,
      "reward_std": 0.30517464876174927,
      "rewards/equation_reward_func": 0.3385416753590107,
      "rewards/format_reward_func": 0.9583333507180214,
      "step": 138
    },
    {
      "completion_length": 440.27344512939453,
      "epoch": 0.07466666666666667,
      "grad_norm": 0.08211855704808459,
      "kl": 0.09521484375,
      "learning_rate": 4.038522720308732e-07,
      "loss": 0.0001,
      "reward": 1.3125000335276127,
      "reward_std": 0.26301832078024745,
      "rewards/equation_reward_func": 0.35416667629033327,
      "rewards/format_reward_func": 0.9583333507180214,
      "step": 140
    },
    {
      "completion_length": 432.4817810058594,
      "epoch": 0.07573333333333333,
      "grad_norm": 0.07820576762133163,
      "kl": 0.098785400390625,
      "learning_rate": 4.009966837786194e-07,
      "loss": 0.0001,
      "reward": 1.367187537252903,
      "reward_std": 0.3508401014842093,
      "rewards/equation_reward_func": 0.4088541781529784,
      "rewards/format_reward_func": 0.9583333544433117,
      "step": 142
    },
    {
      "completion_length": 412.177095413208,
      "epoch": 0.0768,
      "grad_norm": 0.07018309285646165,
      "kl": 0.116241455078125,
      "learning_rate": 3.981097376494259e-07,
      "loss": 0.0001,
      "reward": 1.4244792014360428,
      "reward_std": 0.27811811538413167,
      "rewards/equation_reward_func": 0.45572917629033327,
      "rewards/format_reward_func": 0.9687500260770321,
      "step": 144
    },
    {
      "completion_length": 471.1041736602783,
      "epoch": 0.07786666666666667,
      "grad_norm": 0.07517814151054342,
      "kl": 0.100799560546875,
      "learning_rate": 3.951920331829592e-07,
      "loss": 0.0001,
      "reward": 1.2526041977107525,
      "reward_std": 0.24458088353276253,
      "rewards/equation_reward_func": 0.3020833383779973,
      "rewards/format_reward_func": 0.9505208507180214,
      "step": 146
    },
    {
      "completion_length": 445.95053482055664,
      "epoch": 0.07893333333333333,
      "grad_norm": 0.06913271066183226,
      "kl": 0.0982666015625,
      "learning_rate": 3.922441763065506e-07,
      "loss": 0.0001,
      "reward": 1.3229166939854622,
      "reward_std": 0.2779015698470175,
      "rewards/equation_reward_func": 0.3619791720993817,
      "rewards/format_reward_func": 0.9609375186264515,
      "step": 148
    },
    {
      "completion_length": 426.8359489440918,
      "epoch": 0.08,
      "grad_norm": 0.06291758584225125,
      "kl": 0.107147216796875,
      "learning_rate": 3.8926677920936093e-07,
      "loss": 0.0001,
      "reward": 1.3671875447034836,
      "reward_std": 0.2879944946616888,
      "rewards/equation_reward_func": 0.40364584792405367,
      "rewards/format_reward_func": 0.963541679084301,
      "step": 150
    },
    {
      "completion_length": 475.89323806762695,
      "epoch": 0.08106666666666666,
      "grad_norm": 0.07348149982487694,
      "kl": 0.097625732421875,
      "learning_rate": 3.862604602152464e-07,
      "loss": 0.0001,
      "reward": 1.3229167014360428,
      "reward_std": 0.3060847264714539,
      "rewards/equation_reward_func": 0.3671875102445483,
      "rewards/format_reward_func": 0.955729179084301,
      "step": 152
    },
    {
      "completion_length": 455.48959732055664,
      "epoch": 0.08213333333333334,
      "grad_norm": 0.08190829109797806,
      "kl": 0.10845947265625,
      "learning_rate": 3.8322584365434934e-07,
      "loss": 0.0001,
      "reward": 1.3489583656191826,
      "reward_std": 0.2808792185969651,
      "rewards/equation_reward_func": 0.3828125107102096,
      "rewards/format_reward_func": 0.9661458507180214,
      "step": 154
    },
    {
      "completion_length": 452.7239713668823,
      "epoch": 0.0832,
      "grad_norm": 0.07277189051173198,
      "kl": 0.101318359375,
      "learning_rate": 3.8016355973344173e-07,
      "loss": 0.0001,
      "reward": 1.346354205161333,
      "reward_std": 0.3634042153134942,
      "rewards/equation_reward_func": 0.40625000814907253,
      "rewards/format_reward_func": 0.9401041828095913,
      "step": 156
    },
    {
      "completion_length": 407.30209255218506,
      "epoch": 0.08426666666666667,
      "grad_norm": 0.08347690411933845,
      "kl": 0.105987548828125,
      "learning_rate": 3.7707424440504863e-07,
      "loss": 0.0001,
      "reward": 1.468750037252903,
      "reward_std": 0.25727377785369754,
      "rewards/equation_reward_func": 0.49218750931322575,
      "rewards/format_reward_func": 0.9765625111758709,
      "step": 158
    },
    {
      "completion_length": 471.96095085144043,
      "epoch": 0.08533333333333333,
      "grad_norm": 0.08122691638733275,
      "kl": 0.109375,
      "learning_rate": 3.739585392353787e-07,
      "loss": 0.0001,
      "reward": 1.3151041865348816,
      "reward_std": 0.366784714628011,
      "rewards/equation_reward_func": 0.37239584187045693,
      "rewards/format_reward_func": 0.9427083507180214,
      "step": 160
    },
    {
      "completion_length": 498.70574378967285,
      "epoch": 0.0864,
      "grad_norm": 0.06406400573830406,
      "kl": 0.09796142578125,
      "learning_rate": 3.7081709127108767e-07,
      "loss": 0.0001,
      "reward": 1.1979167014360428,
      "reward_std": 0.24064096482470632,
      "rewards/equation_reward_func": 0.24218750675208867,
      "rewards/format_reward_func": 0.955729179084301,
      "step": 162
    },
    {
      "completion_length": 443.57813453674316,
      "epoch": 0.08746666666666666,
      "grad_norm": 0.08343605647479183,
      "kl": 0.11444091796875,
      "learning_rate": 3.6765055290490513e-07,
      "loss": 0.0001,
      "reward": 1.3750000447034836,
      "reward_std": 0.2543401522561908,
      "rewards/equation_reward_func": 0.4114583474583924,
      "rewards/format_reward_func": 0.9635416865348816,
      "step": 164
    },
    {
      "completion_length": 461.13282585144043,
      "epoch": 0.08853333333333334,
      "grad_norm": 0.07081891374265102,
      "kl": 0.1104736328125,
      "learning_rate": 3.644595817401501e-07,
      "loss": 0.0001,
      "reward": 1.372395858168602,
      "reward_std": 0.28344128327444196,
      "rewards/equation_reward_func": 0.39583334675990045,
      "rewards/format_reward_func": 0.9765625186264515,
      "step": 166
    },
    {
      "completion_length": 406.8073034286499,
      "epoch": 0.0896,
      "grad_norm": 0.06876988951117273,
      "kl": 0.13555908203125,
      "learning_rate": 3.6124484045416483e-07,
      "loss": 0.0001,
      "reward": 1.4661458805203438,
      "reward_std": 0.2285827100276947,
      "rewards/equation_reward_func": 0.4869791753590107,
      "rewards/format_reward_func": 0.9791666865348816,
      "step": 168
    },
    {
      "completion_length": 441.6354293823242,
      "epoch": 0.09066666666666667,
      "grad_norm": 0.05847654146320862,
      "kl": 0.113525390625,
      "learning_rate": 3.580069966606949e-07,
      "loss": 0.0001,
      "reward": 1.3489583805203438,
      "reward_std": 0.22638437105342746,
      "rewards/equation_reward_func": 0.39322918094694614,
      "rewards/format_reward_func": 0.9557291902601719,
      "step": 170
    },
    {
      "completion_length": 396.231782913208,
      "epoch": 0.09173333333333333,
      "grad_norm": 0.08174583936359092,
      "kl": 0.13916015625,
      "learning_rate": 3.547467227712444e-07,
      "loss": 0.0001,
      "reward": 1.4375000335276127,
      "reward_std": 0.24320211308076978,
      "rewards/equation_reward_func": 0.45833334140479565,
      "rewards/format_reward_func": 0.9791666828095913,
      "step": 172
    },
    {
      "completion_length": 469.96876525878906,
      "epoch": 0.0928,
      "grad_norm": 0.07764419704527421,
      "kl": 0.115997314453125,
      "learning_rate": 3.5146469585543386e-07,
      "loss": 0.0001,
      "reward": 1.2734375484287739,
      "reward_std": 0.2251833681948483,
      "rewards/equation_reward_func": 0.3098958428017795,
      "rewards/format_reward_func": 0.9635416865348816,
      "step": 174
    },
    {
      "completion_length": 418.3151168823242,
      "epoch": 0.09386666666666667,
      "grad_norm": 0.07721346430307077,
      "kl": 0.12945556640625,
      "learning_rate": 3.481615975003922e-07,
      "loss": 0.0001,
      "reward": 1.3671875447034836,
      "reward_std": 0.21094412356615067,
      "rewards/equation_reward_func": 0.3880208432674408,
      "rewards/format_reward_func": 0.979166679084301,
      "step": 176
    },
    {
      "completion_length": 420.74740409851074,
      "epoch": 0.09493333333333333,
      "grad_norm": 0.08462924690149998,
      "kl": 0.13238525390625,
      "learning_rate": 3.448381136692089e-07,
      "loss": 0.0001,
      "reward": 1.3489583805203438,
      "reward_std": 0.22936821915209293,
      "rewards/equation_reward_func": 0.3723958458285779,
      "rewards/format_reward_func": 0.9765625186264515,
      "step": 178
    },
    {
      "completion_length": 394.8541784286499,
      "epoch": 0.096,
      "grad_norm": 0.06954554621938241,
      "kl": 0.137664794921875,
      "learning_rate": 3.4149493455847897e-07,
      "loss": 0.0001,
      "reward": 1.4583333656191826,
      "reward_std": 0.22180799348279834,
      "rewards/equation_reward_func": 0.4817708448972553,
      "rewards/format_reward_func": 0.9765625186264515,
      "step": 180
    },
    {
      "completion_length": 410.1119956970215,
      "epoch": 0.09706666666666666,
      "grad_norm": 0.08884342643437457,
      "kl": 0.12933349609375,
      "learning_rate": 3.3813275445496766e-07,
      "loss": 0.0001,
      "reward": 1.3906250223517418,
      "reward_std": 0.2675261232070625,
      "rewards/equation_reward_func": 0.42187501094304025,
      "rewards/format_reward_func": 0.9687500186264515,
      "step": 182
    },
    {
      "completion_length": 425.8515729904175,
      "epoch": 0.09813333333333334,
      "grad_norm": 0.0807749834414267,
      "kl": 0.12939453125,
      "learning_rate": 3.347522715914262e-07,
      "loss": 0.0001,
      "reward": 1.3619792014360428,
      "reward_std": 0.260163695551455,
      "rewards/equation_reward_func": 0.3880208448972553,
      "rewards/format_reward_func": 0.9739583507180214,
      "step": 184
    },
    {
      "completion_length": 414.5390739440918,
      "epoch": 0.0992,
      "grad_norm": 0.0995709776761296,
      "kl": 0.12603759765625,
      "learning_rate": 3.313541880015877e-07,
      "loss": 0.0001,
      "reward": 1.3567708805203438,
      "reward_std": 0.281164659652859,
      "rewards/equation_reward_func": 0.37500001629814506,
      "rewards/format_reward_func": 0.9817708507180214,
      "step": 186
    },
    {
      "completion_length": 434.77605628967285,
      "epoch": 0.10026666666666667,
      "grad_norm": 0.08274171398010909,
      "kl": 0.1339111328125,
      "learning_rate": 3.279392093743747e-07,
      "loss": 0.0001,
      "reward": 1.3046875409781933,
      "reward_std": 0.2186322333291173,
      "rewards/equation_reward_func": 0.3333333448972553,
      "rewards/format_reward_func": 0.9713541865348816,
      "step": 188
    },
    {
      "completion_length": 394.0599060058594,
      "epoch": 0.10133333333333333,
      "grad_norm": 0.09797007100506365,
      "kl": 0.16094970703125,
      "learning_rate": 3.245080449073459e-07,
      "loss": 0.0002,
      "reward": 1.377604216337204,
      "reward_std": 0.24230501474812627,
      "rewards/equation_reward_func": 0.39322918048128486,
      "rewards/format_reward_func": 0.9843750149011612,
      "step": 190
    },
    {
      "completion_length": 360.8489685058594,
      "epoch": 0.1024,
      "grad_norm": 0.08415245014993132,
      "kl": 0.161041259765625,
      "learning_rate": 3.210614071594162e-07,
      "loss": 0.0002,
      "reward": 1.4322917088866234,
      "reward_std": 0.1579188979230821,
      "rewards/equation_reward_func": 0.4505208469927311,
      "rewards/format_reward_func": 0.9817708507180214,
      "step": 192
    },
    {
      "completion_length": 394.7786560058594,
      "epoch": 0.10346666666666667,
      "grad_norm": 0.09090919423012042,
      "kl": 0.15216064453125,
      "learning_rate": 3.1760001190287695e-07,
      "loss": 0.0002,
      "reward": 1.3437500298023224,
      "reward_std": 0.19206074811518192,
      "rewards/equation_reward_func": 0.36979167675599456,
      "rewards/format_reward_func": 0.9739583469927311,
      "step": 194
    },
    {
      "completion_length": 349.3645906448364,
      "epoch": 0.10453333333333334,
      "grad_norm": 0.06511093952525322,
      "kl": 0.163055419921875,
      "learning_rate": 3.141245779747502e-07,
      "loss": 0.0002,
      "reward": 1.4505208730697632,
      "reward_std": 0.20295705134049058,
      "rewards/equation_reward_func": 0.4739583423361182,
      "rewards/format_reward_func": 0.9765625149011612,
      "step": 196
    },
    {
      "completion_length": 347.56771659851074,
      "epoch": 0.1056,
      "grad_norm": 0.12249257016949373,
      "kl": 0.15631103515625,
      "learning_rate": 3.106358271275056e-07,
      "loss": 0.0002,
      "reward": 1.4713542014360428,
      "reward_std": 0.2677521426230669,
      "rewards/equation_reward_func": 0.497395847691223,
      "rewards/format_reward_func": 0.9739583507180214,
      "step": 198
    },
    {
      "completion_length": 353.79949378967285,
      "epoch": 0.10666666666666667,
      "grad_norm": 0.11642634796564656,
      "kl": 0.169464111328125,
      "learning_rate": 3.0713448387917227e-07,
      "loss": 0.0002,
      "reward": 1.424479205161333,
      "reward_std": 0.24417872913181782,
      "rewards/equation_reward_func": 0.455729179084301,
      "rewards/format_reward_func": 0.9687500111758709,
      "step": 200
    },
    {
      "completion_length": 353.62240409851074,
      "epoch": 0.10773333333333333,
      "grad_norm": 0.10966758903292496,
      "kl": 0.17657470703125,
      "learning_rate": 3.0362127536287636e-07,
      "loss": 0.0002,
      "reward": 1.3880208805203438,
      "reward_std": 0.22897254303097725,
      "rewards/equation_reward_func": 0.41145834466442466,
      "rewards/format_reward_func": 0.9765625074505806,
      "step": 202
    },
    {
      "completion_length": 352.95313358306885,
      "epoch": 0.1088,
      "grad_norm": 0.08287713096000299,
      "kl": 0.174285888671875,
      "learning_rate": 3.0009693117583523e-07,
      "loss": 0.0002,
      "reward": 1.4453125298023224,
      "reward_std": 0.23830332793295383,
      "rewards/equation_reward_func": 0.4687500107102096,
      "rewards/format_reward_func": 0.9765625111758709,
      "step": 204
    },
    {
      "completion_length": 342.9895896911621,
      "epoch": 0.10986666666666667,
      "grad_norm": 0.11889984004997693,
      "kl": 0.2039794921875,
      "learning_rate": 2.965621832278401e-07,
      "loss": 0.0002,
      "reward": 1.4661458879709244,
      "reward_std": 0.26223280746489763,
      "rewards/equation_reward_func": 0.4921875160653144,
      "rewards/format_reward_func": 0.9739583544433117,
      "step": 206
    },
    {
      "completion_length": 371.7474021911621,
      "epoch": 0.11093333333333333,
      "grad_norm": 0.08722890145319173,
      "kl": 0.167755126953125,
      "learning_rate": 2.9301776558925875e-07,
      "loss": 0.0002,
      "reward": 1.3437500409781933,
      "reward_std": 0.17667264631018043,
      "rewards/equation_reward_func": 0.380208341171965,
      "rewards/format_reward_func": 0.9635416828095913,
      "step": 208
    },
    {
      "completion_length": 371.341157913208,
      "epoch": 0.112,
      "grad_norm": 0.11118990694564043,
      "kl": 0.1689453125,
      "learning_rate": 2.894644143385885e-07,
      "loss": 0.0002,
      "reward": 1.3880208730697632,
      "reward_std": 0.2672252501361072,
      "rewards/equation_reward_func": 0.41145834187045693,
      "rewards/format_reward_func": 0.9765625111758709,
      "step": 210
    },
    {
      "completion_length": 337.9843807220459,
      "epoch": 0.11306666666666666,
      "grad_norm": 0.0739582977873293,
      "kl": 0.1724853515625,
      "learning_rate": 2.859028674095937e-07,
      "loss": 0.0002,
      "reward": 1.4583333805203438,
      "reward_std": 0.1776336650364101,
      "rewards/equation_reward_func": 0.4739583544433117,
      "rewards/format_reward_func": 0.9843750149011612,
      "step": 212
    },
    {
      "completion_length": 339.21876335144043,
      "epoch": 0.11413333333333334,
      "grad_norm": 0.09143345815857122,
      "kl": 0.17596435546875,
      "learning_rate": 2.823338644380566e-07,
      "loss": 0.0002,
      "reward": 1.492187537252903,
      "reward_std": 0.22370707942172885,
      "rewards/equation_reward_func": 0.5130208460614085,
      "rewards/format_reward_func": 0.9791666828095913,
      "step": 214
    },
    {
      "completion_length": 368.21876430511475,
      "epoch": 0.1152,
      "grad_norm": 0.09155592641067002,
      "kl": 0.19281005859375,
      "learning_rate": 2.7875814660817504e-07,
      "loss": 0.0002,
      "reward": 1.3906250447034836,
      "reward_std": 0.22471351828426123,
      "rewards/equation_reward_func": 0.41666668211109936,
      "rewards/format_reward_func": 0.9739583469927311,
      "step": 216
    },
    {
      "completion_length": 366.52084732055664,
      "epoch": 0.11626666666666667,
      "grad_norm": 0.08581764717856366,
      "kl": 0.17510986328125,
      "learning_rate": 2.751764564986396e-07,
      "loss": 0.0002,
      "reward": 1.4010416939854622,
      "reward_std": 0.2523862696252763,
      "rewards/equation_reward_func": 0.4375000102445483,
      "rewards/format_reward_func": 0.9635416828095913,
      "step": 218
    },
    {
      "completion_length": 382.44011211395264,
      "epoch": 0.11733333333333333,
      "grad_norm": 0.08641987214494386,
      "kl": 0.16357421875,
      "learning_rate": 2.715895379284194e-07,
      "loss": 0.0002,
      "reward": 1.3463542014360428,
      "reward_std": 0.20577920693904161,
      "rewards/equation_reward_func": 0.3776041753590107,
      "rewards/format_reward_func": 0.9687500149011612,
      "step": 220
    },
    {
      "completion_length": 346.23698902130127,
      "epoch": 0.1184,
      "grad_norm": 0.11469342843068384,
      "kl": 0.187591552734375,
      "learning_rate": 2.6799813580229174e-07,
      "loss": 0.0002,
      "reward": 1.414062537252903,
      "reward_std": 0.22339888894930482,
      "rewards/equation_reward_func": 0.4479166741948575,
      "rewards/format_reward_func": 0.9661458507180214,
      "step": 222
    },
    {
      "completion_length": 343.7395906448364,
      "epoch": 0.11946666666666667,
      "grad_norm": 0.10710582787702641,
      "kl": 0.17193603515625,
      "learning_rate": 2.6440299595614606e-07,
      "loss": 0.0002,
      "reward": 1.460937537252903,
      "reward_std": 0.25045704562216997,
      "rewards/equation_reward_func": 0.4921875111758709,
      "rewards/format_reward_func": 0.9687500149011612,
      "step": 224
    },
    {
      "completion_length": 349.78907203674316,
      "epoch": 0.12053333333333334,
      "grad_norm": 0.0754366633012716,
      "kl": 0.180908203125,
      "learning_rate": 2.6080486500209347e-07,
      "loss": 0.0002,
      "reward": 1.421875037252903,
      "reward_std": 0.24725741427391768,
      "rewards/equation_reward_func": 0.46354167629033327,
      "rewards/format_reward_func": 0.9583333507180214,
      "step": 226
    },
    {
      "completion_length": 336.4166774749756,
      "epoch": 0.1216,
      "grad_norm": 0.10961018063074526,
      "kl": 0.1932373046875,
      "learning_rate": 2.572044901734166e-07,
      "loss": 0.0002,
      "reward": 1.4609375521540642,
      "reward_std": 0.24446540977805853,
      "rewards/equation_reward_func": 0.5130208514165133,
      "rewards/format_reward_func": 0.9479166828095913,
      "step": 228
    },
    {
      "completion_length": 362.3151149749756,
      "epoch": 0.12266666666666666,
      "grad_norm": 0.0750432050842772,
      "kl": 0.1806640625,
      "learning_rate": 2.536026191693893e-07,
      "loss": 0.0002,
      "reward": 1.3437500484287739,
      "reward_std": 0.228640450630337,
      "rewards/equation_reward_func": 0.38281250977888703,
      "rewards/format_reward_func": 0.9609375149011612,
      "step": 230
    },
    {
      "completion_length": 346.28646659851074,
      "epoch": 0.12373333333333333,
      "grad_norm": 0.0866455945339987,
      "kl": 0.19073486328125,
      "learning_rate": 2.5e-07,
      "loss": 0.0002,
      "reward": 1.4166667014360428,
      "reward_std": 0.25847903126850724,
      "rewards/equation_reward_func": 0.46875001350417733,
      "rewards/format_reward_func": 0.9479166865348816,
      "step": 232
    },
    {
      "completion_length": 387.15625953674316,
      "epoch": 0.1248,
      "grad_norm": 0.08835289541018657,
      "kl": 0.18060302734375,
      "learning_rate": 2.4639738083061073e-07,
      "loss": 0.0002,
      "reward": 1.3307292126119137,
      "reward_std": 0.2538083433173597,
      "rewards/equation_reward_func": 0.3854166828095913,
      "rewards/format_reward_func": 0.9453125186264515,
      "step": 234
    },
    {
      "completion_length": 371.65625953674316,
      "epoch": 0.12586666666666665,
      "grad_norm": 0.09102684058299942,
      "kl": 0.18902587890625,
      "learning_rate": 2.4279550982658345e-07,
      "loss": 0.0002,
      "reward": 1.369791705161333,
      "reward_std": 0.21475306199863553,
      "rewards/equation_reward_func": 0.4244791795499623,
      "rewards/format_reward_func": 0.9453125186264515,
      "step": 236
    },
    {
      "completion_length": 341.49740505218506,
      "epoch": 0.12693333333333334,
      "grad_norm": 0.08633591554429869,
      "kl": 0.21234130859375,
      "learning_rate": 2.3919513499790646e-07,
      "loss": 0.0002,
      "reward": 1.4531250447034836,
      "reward_std": 0.24704138981178403,
      "rewards/equation_reward_func": 0.5000000195577741,
      "rewards/format_reward_func": 0.9531250149011612,
      "step": 238
    },
    {
      "completion_length": 405.6484489440918,
      "epoch": 0.128,
      "grad_norm": 0.10763242906498482,
      "kl": 0.2197265625,
      "learning_rate": 2.3559700404385394e-07,
      "loss": 0.0002,
      "reward": 1.2890625335276127,
      "reward_std": 0.31284062657505274,
      "rewards/equation_reward_func": 0.3671875048894435,
      "rewards/format_reward_func": 0.9218750223517418,
      "step": 240
    },
    {
      "completion_length": 395.661470413208,
      "epoch": 0.12906666666666666,
      "grad_norm": 0.08038592157957923,
      "kl": 0.2103271484375,
      "learning_rate": 2.3200186419770823e-07,
      "loss": 0.0002,
      "reward": 1.252604205161333,
      "reward_std": 0.2701655197888613,
      "rewards/equation_reward_func": 0.317708337912336,
      "rewards/format_reward_func": 0.934895858168602,
      "step": 242
    },
    {
      "completion_length": 314.59636306762695,
      "epoch": 0.13013333333333332,
      "grad_norm": 0.09192187591314832,
      "kl": 0.22662353515625,
      "learning_rate": 2.284104620715807e-07,
      "loss": 0.0002,
      "reward": 1.4947917088866234,
      "reward_std": 0.3372342623770237,
      "rewards/equation_reward_func": 0.5572916788514704,
      "rewards/format_reward_func": 0.9375000111758709,
      "step": 244
    },
    {
      "completion_length": 329.73698806762695,
      "epoch": 0.1312,
      "grad_norm": 0.10838571395133541,
      "kl": 0.24334716796875,
      "learning_rate": 2.2482354350136043e-07,
      "loss": 0.0002,
      "reward": 1.455729216337204,
      "reward_std": 0.2566018928773701,
      "rewards/equation_reward_func": 0.5104166800156236,
      "rewards/format_reward_func": 0.9453125149011612,
      "step": 246
    },
    {
      "completion_length": 340.03907108306885,
      "epoch": 0.13226666666666667,
      "grad_norm": 0.07566512386282394,
      "kl": 0.2659912109375,
      "learning_rate": 2.2124185339182496e-07,
      "loss": 0.0003,
      "reward": 1.377604205161333,
      "reward_std": 0.2088687182404101,
      "rewards/equation_reward_func": 0.4401041811797768,
      "rewards/format_reward_func": 0.9375000149011612,
      "step": 248
    },
    {
      "completion_length": 391.1614685058594,
      "epoch": 0.13333333333333333,
      "grad_norm": 0.08542314001357776,
      "kl": 0.21875,
      "learning_rate": 2.1766613556194344e-07,
      "loss": 0.0002,
      "reward": 1.2083333767950535,
      "reward_std": 0.24805923039093614,
      "rewards/equation_reward_func": 0.27864584047347307,
      "rewards/format_reward_func": 0.9296875111758709,
      "step": 250
    },
    {
      "completion_length": 338.18751430511475,
      "epoch": 0.1344,
      "grad_norm": 0.08749740784410187,
      "kl": 0.24298095703125,
      "learning_rate": 2.1409713259040628e-07,
      "loss": 0.0002,
      "reward": 1.4062500149011612,
      "reward_std": 0.27646134374663234,
      "rewards/equation_reward_func": 0.4557291779201478,
      "rewards/format_reward_func": 0.9505208507180214,
      "step": 252
    },
    {
      "completion_length": 350.28386306762695,
      "epoch": 0.13546666666666668,
      "grad_norm": 0.10711337890017471,
      "kl": 0.223388671875,
      "learning_rate": 2.105355856614115e-07,
      "loss": 0.0002,
      "reward": 1.3151042014360428,
      "reward_std": 0.25063526490703225,
      "rewards/equation_reward_func": 0.38020834513008595,
      "rewards/format_reward_func": 0.9348958544433117,
      "step": 254
    },
    {
      "completion_length": 339.2656364440918,
      "epoch": 0.13653333333333334,
      "grad_norm": 0.09589698286961762,
      "kl": 0.21533203125,
      "learning_rate": 2.069822344107413e-07,
      "loss": 0.0002,
      "reward": 1.4036458656191826,
      "reward_std": 0.26786494068801403,
      "rewards/equation_reward_func": 0.4453125095460564,
      "rewards/format_reward_func": 0.9583333544433117,
      "step": 256
    },
    {
      "completion_length": 319.2291774749756,
      "epoch": 0.1376,
      "grad_norm": 0.07780726214703354,
      "kl": 0.2373046875,
      "learning_rate": 2.034378167721599e-07,
      "loss": 0.0002,
      "reward": 1.4817708618938923,
      "reward_std": 0.20678498363122344,
      "rewards/equation_reward_func": 0.5234375149011612,
      "rewards/format_reward_func": 0.9583333507180214,
      "step": 258
    },
    {
      "completion_length": 346.63803005218506,
      "epoch": 0.13866666666666666,
      "grad_norm": 0.11641174762994905,
      "kl": 0.22607421875,
      "learning_rate": 1.9990306882416485e-07,
      "loss": 0.0002,
      "reward": 1.3255208656191826,
      "reward_std": 0.24585285549983382,
      "rewards/equation_reward_func": 0.3828125111758709,
      "rewards/format_reward_func": 0.9427083469927311,
      "step": 260
    },
    {
      "completion_length": 337.49740409851074,
      "epoch": 0.13973333333333332,
      "grad_norm": 0.15077694381297332,
      "kl": 0.2508544921875,
      "learning_rate": 1.9637872463712362e-07,
      "loss": 0.0003,
      "reward": 1.4114583656191826,
      "reward_std": 0.3120162379927933,
      "rewards/equation_reward_func": 0.46614584838971496,
      "rewards/format_reward_func": 0.9453125186264515,
      "step": 262
    },
    {
      "completion_length": 350.4427185058594,
      "epoch": 0.1408,
      "grad_norm": 0.09105767195307653,
      "kl": 0.2108154296875,
      "learning_rate": 1.9286551612082773e-07,
      "loss": 0.0002,
      "reward": 1.3541667088866234,
      "reward_std": 0.26035504834726453,
      "rewards/equation_reward_func": 0.4036458460614085,
      "rewards/format_reward_func": 0.9505208507180214,
      "step": 264
    },
    {
      "completion_length": 354.94532108306885,
      "epoch": 0.14186666666666667,
      "grad_norm": 0.11114474779974037,
      "kl": 0.2325439453125,
      "learning_rate": 1.8936417287249446e-07,
      "loss": 0.0002,
      "reward": 1.390625037252903,
      "reward_std": 0.24211309384554625,
      "rewards/equation_reward_func": 0.4296875118743628,
      "rewards/format_reward_func": 0.9609375260770321,
      "step": 266
    },
    {
      "completion_length": 332.6849031448364,
      "epoch": 0.14293333333333333,
      "grad_norm": 0.1123031021076321,
      "kl": 0.229736328125,
      "learning_rate": 1.8587542202524985e-07,
      "loss": 0.0002,
      "reward": 1.3906250409781933,
      "reward_std": 0.2365088271908462,
      "rewards/equation_reward_func": 0.4348958439659327,
      "rewards/format_reward_func": 0.9557291828095913,
      "step": 268
    },
    {
      "completion_length": 345.1770944595337,
      "epoch": 0.144,
      "grad_norm": 0.11158112162119024,
      "kl": 0.21636962890625,
      "learning_rate": 1.82399988097123e-07,
      "loss": 0.0002,
      "reward": 1.361979190260172,
      "reward_std": 0.3079025028273463,
      "rewards/equation_reward_func": 0.43750000395812094,
      "rewards/format_reward_func": 0.9244791865348816,
      "step": 270
    },
    {
      "completion_length": 323.8698034286499,
      "epoch": 0.14506666666666668,
      "grad_norm": 0.09689920100803472,
      "kl": 0.2325439453125,
      "learning_rate": 1.7893859284058378e-07,
      "loss": 0.0002,
      "reward": 1.4296875447034836,
      "reward_std": 0.18631823640316725,
      "rewards/equation_reward_func": 0.46875001303851604,
      "rewards/format_reward_func": 0.9609375186264515,
      "step": 272
    },
    {
      "completion_length": 368.67188262939453,
      "epoch": 0.14613333333333334,
      "grad_norm": 0.13615031599545702,
      "kl": 0.20782470703125,
      "learning_rate": 1.7549195509265407e-07,
      "loss": 0.0002,
      "reward": 1.3203125409781933,
      "reward_std": 0.24673572788015008,
      "rewards/equation_reward_func": 0.3593750079162419,
      "rewards/format_reward_func": 0.9609375186264515,
      "step": 274
    },
    {
      "completion_length": 388.29948806762695,
      "epoch": 0.1472,
      "grad_norm": 0.09073524977124178,
      "kl": 0.20733642578125,
      "learning_rate": 1.7206079062562536e-07,
      "loss": 0.0002,
      "reward": 1.2708333767950535,
      "reward_std": 0.21955121494829655,
      "rewards/equation_reward_func": 0.3281250111758709,
      "rewards/format_reward_func": 0.9427083507180214,
      "step": 276
    },
    {
      "completion_length": 345.38542652130127,
      "epoch": 0.14826666666666666,
      "grad_norm": 5220.06372637961,
      "kl": 7072.191650390625,
      "learning_rate": 1.6864581199841226e-07,
      "loss": 7.0509,
      "reward": 1.3932292088866234,
      "reward_std": 0.25684024626389146,
      "rewards/equation_reward_func": 0.44791668001562357,
      "rewards/format_reward_func": 0.9453125149011612,
      "step": 278
    },
    {
      "completion_length": 319.3724060058594,
      "epoch": 0.14933333333333335,
      "grad_norm": 79.14680202266041,
      "kl": 49.96136474609375,
      "learning_rate": 1.6524772840857388e-07,
      "loss": 0.0497,
      "reward": 1.4244792014360428,
      "reward_std": 0.22301306808367372,
      "rewards/equation_reward_func": 0.46875001303851604,
      "rewards/format_reward_func": 0.9557291902601719,
      "step": 280
    },
    {
      "completion_length": 327.8385486602783,
      "epoch": 0.1504,
      "grad_norm": 0.08745470520702019,
      "kl": 0.24444580078125,
      "learning_rate": 1.6186724554503237e-07,
      "loss": 0.0002,
      "reward": 1.4192708693444729,
      "reward_std": 0.22730597108602524,
      "rewards/equation_reward_func": 0.45833334093913436,
      "rewards/format_reward_func": 0.9609375186264515,
      "step": 282
    },
    {
      "completion_length": 321.70053005218506,
      "epoch": 0.15146666666666667,
      "grad_norm": 0.09168203339141828,
      "kl": 0.20928955078125,
      "learning_rate": 1.5850506544152103e-07,
      "loss": 0.0002,
      "reward": 1.4739583656191826,
      "reward_std": 0.19128448516130447,
      "rewards/equation_reward_func": 0.5052083432674408,
      "rewards/format_reward_func": 0.9687500149011612,
      "step": 284
    },
    {
      "completion_length": 322.1458406448364,
      "epoch": 0.15253333333333333,
      "grad_norm": 0.12387430706161645,
      "kl": 0.21209716796875,
      "learning_rate": 1.5516188633079107e-07,
      "loss": 0.0002,
      "reward": 1.4322917088866234,
      "reward_std": 0.2435653922148049,
      "rewards/equation_reward_func": 0.4817708421032876,
      "rewards/format_reward_func": 0.9505208507180214,
      "step": 286
    },
    {
      "completion_length": 335.86459159851074,
      "epoch": 0.1536,
      "grad_norm": 0.11042480316797151,
      "kl": 0.2279052734375,
      "learning_rate": 1.5183840249960784e-07,
      "loss": 0.0002,
      "reward": 1.4062500409781933,
      "reward_std": 0.2505862903781235,
      "rewards/equation_reward_func": 0.4791666779201478,
      "rewards/format_reward_func": 0.9270833507180214,
      "step": 288
    },
    {
      "completion_length": 332.4531373977661,
      "epoch": 0.15466666666666667,
      "grad_norm": 0.11068445556167467,
      "kl": 0.226318359375,
      "learning_rate": 1.4853530414456612e-07,
      "loss": 0.0002,
      "reward": 1.3932291939854622,
      "reward_std": 0.2191076222807169,
      "rewards/equation_reward_func": 0.44791667349636555,
      "rewards/format_reward_func": 0.9453125149011612,
      "step": 290
    },
    {
      "completion_length": 346.98959159851074,
      "epoch": 0.15573333333333333,
      "grad_norm": 0.10537288030437299,
      "kl": 0.2379150390625,
      "learning_rate": 1.4525327722875568e-07,
      "loss": 0.0002,
      "reward": 1.367187537252903,
      "reward_std": 0.25823584850877523,
      "rewards/equation_reward_func": 0.4270833458285779,
      "rewards/format_reward_func": 0.9401041828095913,
      "step": 292
    },
    {
      "completion_length": 325.7448043823242,
      "epoch": 0.1568,
      "grad_norm": 0.10313889262076706,
      "kl": 0.21075439453125,
      "learning_rate": 1.4199300333930515e-07,
      "loss": 0.0002,
      "reward": 1.390625037252903,
      "reward_std": 0.24558233935385942,
      "rewards/equation_reward_func": 0.4505208465270698,
      "rewards/format_reward_func": 0.9401041865348816,
      "step": 294
    },
    {
      "completion_length": 324.1849069595337,
      "epoch": 0.15786666666666666,
      "grad_norm": 0.10081124744539481,
      "kl": 0.23394775390625,
      "learning_rate": 1.3875515954583523e-07,
      "loss": 0.0002,
      "reward": 1.3984375447034836,
      "reward_std": 0.24683487368747592,
      "rewards/equation_reward_func": 0.4505208465270698,
      "rewards/format_reward_func": 0.9479166828095913,
      "step": 296
    },
    {
      "completion_length": 321.01303005218506,
      "epoch": 0.15893333333333334,
      "grad_norm": 0.08166984314299437,
      "kl": 0.21270751953125,
      "learning_rate": 1.3554041825985e-07,
      "loss": 0.0002,
      "reward": 1.4427083656191826,
      "reward_std": 0.15349334990605712,
      "rewards/equation_reward_func": 0.48177084513008595,
      "rewards/format_reward_func": 0.9609375223517418,
      "step": 298
    },
    {
      "completion_length": 326.8463659286499,
      "epoch": 0.16,
      "grad_norm": 0.07831370889731273,
      "kl": 0.23455810546875,
      "learning_rate": 1.323494470950949e-07,
      "loss": 0.0002,
      "reward": 1.351562537252903,
      "reward_std": 0.22833695402368903,
      "rewards/equation_reward_func": 0.4036458421032876,
      "rewards/format_reward_func": 0.9479166865348816,
      "step": 300
    },
    {
      "completion_length": 344.11719512939453,
      "epoch": 0.16106666666666666,
      "grad_norm": 0.08232548207891384,
      "kl": 0.22637939453125,
      "learning_rate": 1.2918290872891236e-07,
      "loss": 0.0002,
      "reward": 1.377604205161333,
      "reward_std": 0.2291308455169201,
      "rewards/equation_reward_func": 0.42968751303851604,
      "rewards/format_reward_func": 0.9479166865348816,
      "step": 302
    },
    {
      "completion_length": 326.5494899749756,
      "epoch": 0.16213333333333332,
      "grad_norm": 0.10750065685022393,
      "kl": 0.21270751953125,
      "learning_rate": 1.260414607646213e-07,
      "loss": 0.0002,
      "reward": 1.408854205161333,
      "reward_std": 0.24541507568210363,
      "rewards/equation_reward_func": 0.45572917931713164,
      "rewards/format_reward_func": 0.9531250149011612,
      "step": 304
    },
    {
      "completion_length": 310.8593854904175,
      "epoch": 0.1632,
      "grad_norm": 0.0890480361990402,
      "kl": 0.2191162109375,
      "learning_rate": 1.2292575559495143e-07,
      "loss": 0.0002,
      "reward": 1.455729205161333,
      "reward_std": 0.18851999705657363,
      "rewards/equation_reward_func": 0.5104166793171316,
      "rewards/format_reward_func": 0.9453125223517418,
      "step": 306
    },
    {
      "completion_length": 336.3177185058594,
      "epoch": 0.16426666666666667,
      "grad_norm": 0.12539643357527022,
      "kl": 0.22491455078125,
      "learning_rate": 1.1983644026655835e-07,
      "loss": 0.0002,
      "reward": 1.398437526077032,
      "reward_std": 0.24129127291962504,
      "rewards/equation_reward_func": 0.43489584792405367,
      "rewards/format_reward_func": 0.963541679084301,
      "step": 308
    },
    {
      "completion_length": 363.4453248977661,
      "epoch": 0.16533333333333333,
      "grad_norm": 0.10147317845374816,
      "kl": 0.22625732421875,
      "learning_rate": 1.1677415634565066e-07,
      "loss": 0.0002,
      "reward": 1.2890625298023224,
      "reward_std": 0.2331953472457826,
      "rewards/equation_reward_func": 0.3359375095460564,
      "rewards/format_reward_func": 0.9531250186264515,
      "step": 310
    },
    {
      "completion_length": 334.07552909851074,
      "epoch": 0.1664,
      "grad_norm": 0.07071630516660898,
      "kl": 0.21331787109375,
      "learning_rate": 1.1373953978475353e-07,
      "loss": 0.0002,
      "reward": 1.3776042088866234,
      "reward_std": 0.2111625811085105,
      "rewards/equation_reward_func": 0.42708334140479565,
      "rewards/format_reward_func": 0.9505208469927311,
      "step": 312
    },
    {
      "completion_length": 345.9349031448364,
      "epoch": 0.16746666666666668,
      "grad_norm": 0.09304504389856937,
      "kl": 0.22332763671875,
      "learning_rate": 1.1073322079063913e-07,
      "loss": 0.0002,
      "reward": 1.3802083693444729,
      "reward_std": 0.24213288770988584,
      "rewards/equation_reward_func": 0.4453125111758709,
      "rewards/format_reward_func": 0.9348958507180214,
      "step": 314
    },
    {
      "completion_length": 343.2890691757202,
      "epoch": 0.16853333333333334,
      "grad_norm": 0.1097380319531818,
      "kl": 0.22332763671875,
      "learning_rate": 1.0775582369344946e-07,
      "loss": 0.0002,
      "reward": 1.3098958730697632,
      "reward_std": 0.29304072819650173,
      "rewards/equation_reward_func": 0.3776041774544865,
      "rewards/format_reward_func": 0.9322916828095913,
      "step": 316
    },
    {
      "completion_length": 294.8020906448364,
      "epoch": 0.1696,
      "grad_norm": 1.43305351784789,
      "kl": 3.03558349609375,
      "learning_rate": 1.0480796681704077e-07,
      "loss": 0.003,
      "reward": 1.5338542014360428,
      "reward_std": 0.18905250960960984,
      "rewards/equation_reward_func": 0.5651041741948575,
      "rewards/format_reward_func": 0.9687500186264515,
      "step": 318
    },
    {
      "completion_length": 301.0208396911621,
      "epoch": 0.17066666666666666,
      "grad_norm": 0.09915602747223792,
      "kl": 0.23931884765625,
      "learning_rate": 1.018902623505741e-07,
      "loss": 0.0002,
      "reward": 1.5182292088866234,
      "reward_std": 0.19097376568242908,
      "rewards/equation_reward_func": 0.5625000158324838,
      "rewards/format_reward_func": 0.9557291902601719,
      "step": 320
    },
    {
      "completion_length": 320.85417461395264,
      "epoch": 0.17173333333333332,
      "grad_norm": 0.09117919942079071,
      "kl": 0.214111328125,
      "learning_rate": 9.900331622138063e-08,
      "loss": 0.0002,
      "reward": 1.4140625521540642,
      "reward_std": 0.2838053968735039,
      "rewards/equation_reward_func": 0.460937516996637,
      "rewards/format_reward_func": 0.9531250223517418,
      "step": 322
    },
    {
      "completion_length": 311.2395896911621,
      "epoch": 0.1728,
      "grad_norm": 0.0815049784953199,
      "kl": 0.22808837890625,
      "learning_rate": 9.614772796912681e-08,
      "loss": 0.0002,
      "reward": 1.453125037252903,
      "reward_std": 0.2417630972340703,
      "rewards/equation_reward_func": 0.5182291767559946,
      "rewards/format_reward_func": 0.9348958469927311,
      "step": 324
    },
    {
      "completion_length": 333.16146755218506,
      "epoch": 0.17386666666666667,
      "grad_norm": 0.10328198816513103,
      "kl": 0.22216796875,
      "learning_rate": 9.332409062130686e-08,
      "loss": 0.0002,
      "reward": 1.3906250335276127,
      "reward_std": 0.237897964194417,
      "rewards/equation_reward_func": 0.43750001583248377,
      "rewards/format_reward_func": 0.9531250149011612,
      "step": 326
    },
    {
      "completion_length": 310.9401111602783,
      "epoch": 0.17493333333333333,
      "grad_norm": 0.11902359297904073,
      "kl": 0.2431640625,
      "learning_rate": 9.053299057008699e-08,
      "loss": 0.0002,
      "reward": 1.4453125409781933,
      "reward_std": 0.26797239342704415,
      "rewards/equation_reward_func": 0.5000000176951289,
      "rewards/format_reward_func": 0.9453125111758709,
      "step": 328
    },
    {
      "completion_length": 290.20573902130127,
      "epoch": 0.176,
      "grad_norm": 0.0870223579021212,
      "kl": 0.244873046875,
      "learning_rate": 8.777500745052743e-08,
      "loss": 0.0002,
      "reward": 1.5052083805203438,
      "reward_std": 0.21289193397387862,
      "rewards/equation_reward_func": 0.5468750149011612,
      "rewards/format_reward_func": 0.9583333469927311,
      "step": 330
    },
    {
      "completion_length": 317.96094512939453,
      "epoch": 0.17706666666666668,
      "grad_norm": 0.09923025666525324,
      "kl": 0.24334716796875,
      "learning_rate": 8.505071402020892e-08,
      "loss": 0.0002,
      "reward": 1.3723958656191826,
      "reward_std": 0.19928161799907684,
      "rewards/equation_reward_func": 0.42187501303851604,
      "rewards/format_reward_func": 0.9505208507180214,
      "step": 332
    },
    {
      "completion_length": 282.44271659851074,
      "epoch": 0.17813333333333334,
      "grad_norm": 0.11232051976601577,
      "kl": 0.25335693359375,
      "learning_rate": 8.236067604028562e-08,
      "loss": 0.0003,
      "reward": 1.5130208656191826,
      "reward_std": 0.24858005391433835,
      "rewards/equation_reward_func": 0.5546875162981451,
      "rewards/format_reward_func": 0.9583333544433117,
      "step": 334
    },
    {
      "completion_length": 318.4244842529297,
      "epoch": 0.1792,
      "grad_norm": 0.1083015800637841,
      "kl": 0.213134765625,
      "learning_rate": 7.970545215799327e-08,
      "loss": 0.0002,
      "reward": 1.4739583730697632,
      "reward_std": 0.2625010758638382,
      "rewards/equation_reward_func": 0.531250013038516,
      "rewards/format_reward_func": 0.9427083507180214,
      "step": 336
    },
    {
      "completion_length": 340.153657913208,
      "epoch": 0.18026666666666666,
      "grad_norm": 0.09142278500214475,
      "kl": 0.23583984375,
      "learning_rate": 7.708559379063204e-08,
      "loss": 0.0002,
      "reward": 1.406250037252903,
      "reward_std": 0.22931140661239624,
      "rewards/equation_reward_func": 0.4453125102445483,
      "rewards/format_reward_func": 0.9609375111758709,
      "step": 338
    },
    {
      "completion_length": 316.268235206604,
      "epoch": 0.18133333333333335,
      "grad_norm": 0.08608717054817108,
      "kl": 0.2203369140625,
      "learning_rate": 7.45016450110534e-08,
      "loss": 0.0002,
      "reward": 1.3828125335276127,
      "reward_std": 0.1976271690800786,
      "rewards/equation_reward_func": 0.43229167512618005,
      "rewards/format_reward_func": 0.950520858168602,
      "step": 340
    },
    {
      "completion_length": 332.22657108306885,
      "epoch": 0.1824,
      "grad_norm": 0.08626389120583622,
      "kl": 0.22198486328125,
      "learning_rate": 7.195414243467029e-08,
      "loss": 0.0002,
      "reward": 1.3906250409781933,
      "reward_std": 0.20977461012080312,
      "rewards/equation_reward_func": 0.45833334675990045,
      "rewards/format_reward_func": 0.9322916865348816,
      "step": 342
    },
    {
      "completion_length": 321.02605152130127,
      "epoch": 0.18346666666666667,
      "grad_norm": 0.11346266373826759,
      "kl": 0.210693359375,
      "learning_rate": 6.944361510801763e-08,
      "loss": 0.0002,
      "reward": 1.4661458730697632,
      "reward_std": 0.2054482251405716,
      "rewards/equation_reward_func": 0.4973958432674408,
      "rewards/format_reward_func": 0.9687500149011612,
      "step": 344
    },
    {
      "completion_length": 298.59636211395264,
      "epoch": 0.18453333333333333,
      "grad_norm": 0.1067949932858023,
      "kl": 0.29071044921875,
      "learning_rate": 6.697058439888283e-08,
      "loss": 0.0003,
      "reward": 1.390625037252903,
      "reward_std": 0.24251853488385677,
      "rewards/equation_reward_func": 0.4531250107102096,
      "rewards/format_reward_func": 0.9375000223517418,
      "step": 346
    },
    {
      "completion_length": 323.62761211395264,
      "epoch": 0.1856,
      "grad_norm": 0.10153580772849999,
      "kl": 0.20916748046875,
      "learning_rate": 6.453556388803288e-08,
      "loss": 0.0002,
      "reward": 1.3697917088866234,
      "reward_std": 0.20808159746229649,
      "rewards/equation_reward_func": 0.42447917722165585,
      "rewards/format_reward_func": 0.9453125223517418,
      "step": 348
    },
    {
      "completion_length": 309.18751096725464,
      "epoch": 0.18666666666666668,
      "grad_norm": 0.08506346440717698,
      "kl": 0.2176513671875,
      "learning_rate": 6.213905926255697e-08,
      "loss": 0.0002,
      "reward": 1.3958333693444729,
      "reward_std": 0.16924588894471526,
      "rewards/equation_reward_func": 0.43229168001562357,
      "rewards/format_reward_func": 0.9635416828095913,
      "step": 350
    },
    {
      "completion_length": 276.46875858306885,
      "epoch": 0.18773333333333334,
      "grad_norm": 0.07380858754669188,
      "kl": 0.23443603515625,
      "learning_rate": 5.978156821084987e-08,
      "loss": 0.0002,
      "reward": 1.5572916865348816,
      "reward_std": 0.1799696683883667,
      "rewards/equation_reward_func": 0.5781250111758709,
      "rewards/format_reward_func": 0.9791666828095913,
      "step": 352
    },
    {
      "completion_length": 317.0286521911621,
      "epoch": 0.1888,
      "grad_norm": 0.08221686789004773,
      "kl": 0.221435546875,
      "learning_rate": 5.7463580319254853e-08,
      "loss": 0.0002,
      "reward": 1.4348958693444729,
      "reward_std": 0.18001093342900276,
      "rewards/equation_reward_func": 0.4713541779201478,
      "rewards/format_reward_func": 0.9635416828095913,
      "step": 354
    },
    {
      "completion_length": 341.17709255218506,
      "epoch": 0.18986666666666666,
      "grad_norm": 0.1006850259811907,
      "kl": 0.21112060546875,
      "learning_rate": 5.518557697039081e-08,
      "loss": 0.0002,
      "reward": 1.3802083730697632,
      "reward_std": 0.24440749734640121,
      "rewards/equation_reward_func": 0.4322916748933494,
      "rewards/format_reward_func": 0.9479166865348816,
      "step": 356
    },
    {
      "completion_length": 282.05209159851074,
      "epoch": 0.19093333333333334,
      "grad_norm": 0.0784480555352209,
      "kl": 0.23577880859375,
      "learning_rate": 5.294803124318145e-08,
      "loss": 0.0002,
      "reward": 1.5859375223517418,
      "reward_std": 0.19743689289316535,
      "rewards/equation_reward_func": 0.6302083439659327,
      "rewards/format_reward_func": 0.9557291828095913,
      "step": 358
    },
    {
      "completion_length": 312.36459255218506,
      "epoch": 0.192,
      "grad_norm": 0.09968321821935404,
      "kl": 0.2149658203125,
      "learning_rate": 5.07514078146106e-08,
      "loss": 0.0002,
      "reward": 1.3828125409781933,
      "reward_std": 0.23581918003037572,
      "rewards/equation_reward_func": 0.42708334885537624,
      "rewards/format_reward_func": 0.9557291828095913,
      "step": 360
    },
    {
      "completion_length": 310.42968940734863,
      "epoch": 0.19306666666666666,
      "grad_norm": 0.08047116966625914,
      "kl": 0.2113037109375,
      "learning_rate": 4.859616286322094e-08,
      "loss": 0.0002,
      "reward": 1.5208333730697632,
      "reward_std": 0.18892288114875555,
      "rewards/equation_reward_func": 0.5625000111758709,
      "rewards/format_reward_func": 0.9583333469927311,
      "step": 362
    },
    {
      "completion_length": 334.89063262939453,
      "epoch": 0.19413333333333332,
      "grad_norm": 0.11871865226917534,
      "kl": 0.22442626953125,
      "learning_rate": 4.648274397437829e-08,
      "loss": 0.0002,
      "reward": 1.3359375186264515,
      "reward_std": 0.24623532965779305,
      "rewards/equation_reward_func": 0.3984375074505806,
      "rewards/format_reward_func": 0.9375000149011612,
      "step": 364
    },
    {
      "completion_length": 321.46875953674316,
      "epoch": 0.1952,
      "grad_norm": 0.08169405576123293,
      "kl": 0.21612548828125,
      "learning_rate": 4.4411590047320617e-08,
      "loss": 0.0002,
      "reward": 1.4192708805203438,
      "reward_std": 0.1580515131354332,
      "rewards/equation_reward_func": 0.45833334466442466,
      "rewards/format_reward_func": 0.9609375223517418,
      "step": 366
    },
    {
      "completion_length": 268.88021659851074,
      "epoch": 0.19626666666666667,
      "grad_norm": 0.09213898057952202,
      "kl": 0.2569580078125,
      "learning_rate": 4.2383131204010494e-08,
      "loss": 0.0003,
      "reward": 1.5286458656191826,
      "reward_std": 0.19515367224812508,
      "rewards/equation_reward_func": 0.5677083488553762,
      "rewards/format_reward_func": 0.9609375186264515,
      "step": 368
    },
    {
      "completion_length": 308.2838611602783,
      "epoch": 0.19733333333333333,
      "grad_norm": 0.102066920867919,
      "kl": 0.2474365234375,
      "learning_rate": 4.039778869981064e-08,
      "loss": 0.0002,
      "reward": 1.424479205161333,
      "reward_std": 0.16935323411598802,
      "rewards/equation_reward_func": 0.45312500931322575,
      "rewards/format_reward_func": 0.971354179084301,
      "step": 370
    },
    {
      "completion_length": 328.89323806762695,
      "epoch": 0.1984,
      "grad_norm": 0.10475529028007942,
      "kl": 0.218505859375,
      "learning_rate": 3.845597483600049e-08,
      "loss": 0.0002,
      "reward": 1.3697917014360428,
      "reward_std": 0.21701573207974434,
      "rewards/equation_reward_func": 0.42708334466442466,
      "rewards/format_reward_func": 0.9427083544433117,
      "step": 372
    },
    {
      "completion_length": 336.2552156448364,
      "epoch": 0.19946666666666665,
      "grad_norm": 0.0856518942239391,
      "kl": 0.2081298828125,
      "learning_rate": 3.655809287415284e-08,
      "loss": 0.0002,
      "reward": 1.3567708916962147,
      "reward_std": 0.1945798760280013,
      "rewards/equation_reward_func": 0.40364584513008595,
      "rewards/format_reward_func": 0.9531250260770321,
      "step": 374
    },
    {
      "completion_length": 341.63542652130127,
      "epoch": 0.20053333333333334,
      "grad_norm": 0.1246882559841069,
      "kl": 0.21630859375,
      "learning_rate": 3.4704536952387285e-08,
      "loss": 0.0002,
      "reward": 1.3333333656191826,
      "reward_std": 0.23673815419897437,
      "rewards/equation_reward_func": 0.3723958423361182,
      "rewards/format_reward_func": 0.9609375186264515,
      "step": 376
    },
    {
      "completion_length": 306.4270896911621,
      "epoch": 0.2016,
      "grad_norm": 0.1209625715448826,
      "kl": 0.23431396484375,
      "learning_rate": 3.2895692003518575e-08,
      "loss": 0.0002,
      "reward": 1.3958333805203438,
      "reward_std": 0.22822495363652706,
      "rewards/equation_reward_func": 0.44531251303851604,
      "rewards/format_reward_func": 0.9505208544433117,
      "step": 378
    },
    {
      "completion_length": 309.70313453674316,
      "epoch": 0.20266666666666666,
      "grad_norm": 0.12006215018778782,
      "kl": 0.2227783203125,
      "learning_rate": 3.113193367511635e-08,
      "loss": 0.0002,
      "reward": 1.468750037252903,
      "reward_std": 0.23679165728390217,
      "rewards/equation_reward_func": 0.5208333502523601,
      "rewards/format_reward_func": 0.9479166828095913,
      "step": 380
    },
    {
      "completion_length": 317.74740409851074,
      "epoch": 0.20373333333333332,
      "grad_norm": 0.10748128632252552,
      "kl": 0.21893310546875,
      "learning_rate": 2.9413628251493934e-08,
      "loss": 0.0002,
      "reward": 1.4244792088866234,
      "reward_std": 0.197633049916476,
      "rewards/equation_reward_func": 0.458333347691223,
      "rewards/format_reward_func": 0.9661458544433117,
      "step": 382
    },
    {
      "completion_length": 300.8619918823242,
      "epoch": 0.2048,
      "grad_norm": 0.09907271619590914,
      "kl": 0.31658935546875,
      "learning_rate": 2.774113257764066e-08,
      "loss": 0.0003,
      "reward": 1.486979216337204,
      "reward_std": 0.262174597941339,
      "rewards/equation_reward_func": 0.536458345130086,
      "rewards/format_reward_func": 0.950520858168602,
      "step": 384
    },
    {
      "completion_length": 326.85417461395264,
      "epoch": 0.20586666666666667,
      "grad_norm": 0.07810355072932347,
      "kl": 0.22857666015625,
      "learning_rate": 2.611479398511518e-08,
      "loss": 0.0002,
      "reward": 1.322916705161333,
      "reward_std": 0.17400957271456718,
      "rewards/equation_reward_func": 0.3750000100117177,
      "rewards/format_reward_func": 0.9479166865348816,
      "step": 386
    },
    {
      "completion_length": 299.80730056762695,
      "epoch": 0.20693333333333333,
      "grad_norm": 0.10017158729015543,
      "kl": 0.22186279296875,
      "learning_rate": 2.4534950219914057e-08,
      "loss": 0.0002,
      "reward": 1.4869792088866234,
      "reward_std": 0.23186511639505625,
      "rewards/equation_reward_func": 0.5286458488553762,
      "rewards/format_reward_func": 0.958333358168602,
      "step": 388
    },
    {
      "completion_length": 320.9270896911621,
      "epoch": 0.208,
      "grad_norm": 0.1099683165292792,
      "kl": 0.21923828125,
      "learning_rate": 2.300192937233128e-08,
      "loss": 0.0002,
      "reward": 1.4114583618938923,
      "reward_std": 0.22744439914822578,
      "rewards/equation_reward_func": 0.460937513737008,
      "rewards/format_reward_func": 0.9505208544433117,
      "step": 390
    },
    {
      "completion_length": 325.35417652130127,
      "epoch": 0.20906666666666668,
      "grad_norm": 0.0949023163219697,
      "kl": 0.2340087890625,
      "learning_rate": 2.1516049808822935e-08,
      "loss": 0.0002,
      "reward": 1.3906250484287739,
      "reward_std": 0.22641806630417705,
      "rewards/equation_reward_func": 0.44270834838971496,
      "rewards/format_reward_func": 0.9479166828095913,
      "step": 392
    },
    {
      "completion_length": 297.90365505218506,
      "epoch": 0.21013333333333334,
      "grad_norm": 0.1317731635826343,
      "kl": 0.22271728515625,
      "learning_rate": 2.007762010589098e-08,
      "loss": 0.0002,
      "reward": 1.5130208693444729,
      "reward_std": 0.25085952738299966,
      "rewards/equation_reward_func": 0.5520833525806665,
      "rewards/format_reward_func": 0.9609375186264515,
      "step": 394
    },
    {
      "completion_length": 308.4817838668823,
      "epoch": 0.2112,
      "grad_norm": 0.18333770809820993,
      "kl": 0.3961181640625,
      "learning_rate": 1.8686938986000627e-08,
      "loss": 0.0004,
      "reward": 1.377604205161333,
      "reward_std": 0.16871334984898567,
      "rewards/equation_reward_func": 0.416666679084301,
      "rewards/format_reward_func": 0.9609375186264515,
      "step": 396
    },
    {
      "completion_length": 323.61198902130127,
      "epoch": 0.21226666666666666,
      "grad_norm": 0.10822818670117657,
      "kl": 0.22650146484375,
      "learning_rate": 1.734429525554365e-08,
      "loss": 0.0002,
      "reward": 1.328125037252903,
      "reward_std": 0.22037217440083623,
      "rewards/equation_reward_func": 0.3828125118743628,
      "rewards/format_reward_func": 0.9453125223517418,
      "step": 398
    },
    {
      "completion_length": 310.49480152130127,
      "epoch": 0.21333333333333335,
      "grad_norm": 0.08677594432490206,
      "kl": 0.2261962890625,
      "learning_rate": 1.604996774486145e-08,
      "loss": 0.0002,
      "reward": 1.3958333730697632,
      "reward_std": 0.16300127794966102,
      "rewards/equation_reward_func": 0.44270834466442466,
      "rewards/format_reward_func": 0.9531250186264515,
      "step": 400
    },
    {
      "completion_length": 315.83594608306885,
      "epoch": 0.2144,
      "grad_norm": 0.09329093411722503,
      "kl": 0.21697998046875,
      "learning_rate": 1.4804225250339281e-08,
      "loss": 0.0002,
      "reward": 1.4583333656191826,
      "reward_std": 0.2248752093873918,
      "rewards/equation_reward_func": 0.4947916781529784,
      "rewards/format_reward_func": 0.9635416828095913,
      "step": 402
    },
    {
      "completion_length": 305.27605056762695,
      "epoch": 0.21546666666666667,
      "grad_norm": 0.09309626955656056,
      "kl": 0.22406005859375,
      "learning_rate": 1.360732647858498e-08,
      "loss": 0.0002,
      "reward": 1.4114583805203438,
      "reward_std": 0.25712150149047375,
      "rewards/equation_reward_func": 0.46093751303851604,
      "rewards/format_reward_func": 0.9505208507180214,
      "step": 404
    },
    {
      "completion_length": 300.04948902130127,
      "epoch": 0.21653333333333333,
      "grad_norm": 0.06996801120635364,
      "kl": 0.22509765625,
      "learning_rate": 1.2459519992702311e-08,
      "loss": 0.0002,
      "reward": 1.4218750409781933,
      "reward_std": 0.15984355006366968,
      "rewards/equation_reward_func": 0.4583333460614085,
      "rewards/format_reward_func": 0.9635416753590107,
      "step": 406
    },
    {
      "completion_length": 353.1041774749756,
      "epoch": 0.2176,
      "grad_norm": 0.08583468471450732,
      "kl": 0.231201171875,
      "learning_rate": 1.1361044160671629e-08,
      "loss": 0.0002,
      "reward": 1.3619791977107525,
      "reward_std": 0.21500447858124971,
      "rewards/equation_reward_func": 0.41145834792405367,
      "rewards/format_reward_func": 0.9505208544433117,
      "step": 408
    },
    {
      "completion_length": 315.5390739440918,
      "epoch": 0.21866666666666668,
      "grad_norm": 0.10411756903686771,
      "kl": 0.20782470703125,
      "learning_rate": 1.0312127105846947e-08,
      "loss": 0.0002,
      "reward": 1.3854167088866234,
      "reward_std": 0.1989023443311453,
      "rewards/equation_reward_func": 0.41666667722165585,
      "rewards/format_reward_func": 0.9687500149011612,
      "step": 410
    },
    {
      "completion_length": 289.3567810058594,
      "epoch": 0.21973333333333334,
      "grad_norm": 0.08432337279172568,
      "kl": 0.2266845703125,
      "learning_rate": 9.312986659581301e-09,
      "loss": 0.0002,
      "reward": 1.5078125409781933,
      "reward_std": 0.16473140195012093,
      "rewards/equation_reward_func": 0.5286458446644247,
      "rewards/format_reward_func": 0.979166679084301,
      "step": 412
    },
    {
      "completion_length": 328.8099031448364,
      "epoch": 0.2208,
      "grad_norm": 0.08818627291080708,
      "kl": 0.21533203125,
      "learning_rate": 8.363830315988945e-09,
      "loss": 0.0002,
      "reward": 1.403645858168602,
      "reward_std": 0.20907884137704968,
      "rewards/equation_reward_func": 0.4531250139698386,
      "rewards/format_reward_func": 0.9505208507180214,
      "step": 414
    },
    {
      "completion_length": 294.4974031448364,
      "epoch": 0.22186666666666666,
      "grad_norm": 0.09531400720095902,
      "kl": 0.2193603515625,
      "learning_rate": 7.46485518885462e-09,
      "loss": 0.0002,
      "reward": 1.442708384245634,
      "reward_std": 0.18354794522747397,
      "rewards/equation_reward_func": 0.4791666781529784,
      "rewards/format_reward_func": 0.9635416828095913,
      "step": 416
    },
    {
      "completion_length": 324.81511211395264,
      "epoch": 0.22293333333333334,
      "grad_norm": 0.1078543321671686,
      "kl": 0.2205810546875,
      "learning_rate": 6.616247970698319e-09,
      "loss": 0.0002,
      "reward": 1.4036458656191826,
      "reward_std": 0.20324593503028154,
      "rewards/equation_reward_func": 0.42968750931322575,
      "rewards/format_reward_func": 0.9739583469927311,
      "step": 418
    },
    {
      "completion_length": 329.2942781448364,
      "epoch": 0.224,
      "grad_norm": 0.07753331154087116,
      "kl": 0.2059326171875,
      "learning_rate": 5.8181848940044855e-09,
      "loss": 0.0002,
      "reward": 1.343750037252903,
      "reward_std": 0.20026329206302762,
      "rewards/equation_reward_func": 0.39322917629033327,
      "rewards/format_reward_func": 0.9505208469927311,
      "step": 420
    },
    {
      "completion_length": 258.8515682220459,
      "epoch": 0.22506666666666666,
      "grad_norm": 1.0694633359031593,
      "kl": 1.15753173828125,
      "learning_rate": 5.070831694623135e-09,
      "loss": 0.0012,
      "reward": 1.575520858168602,
      "reward_std": 0.1907555889338255,
      "rewards/equation_reward_func": 0.6197916809469461,
      "rewards/format_reward_func": 0.9557291828095913,
      "step": 422
    },
    {
      "completion_length": 340.5260534286499,
      "epoch": 0.22613333333333333,
      "grad_norm": 0.13996783116669323,
      "kl": 0.2117919921875,
      "learning_rate": 4.374343577351336e-09,
      "loss": 0.0002,
      "reward": 1.3437500596046448,
      "reward_std": 0.21045360853895545,
      "rewards/equation_reward_func": 0.398437513737008,
      "rewards/format_reward_func": 0.9453125149011612,
      "step": 424
    },
    {
      "completion_length": 314.54167461395264,
      "epoch": 0.2272,
      "grad_norm": 0.11553621912416584,
      "kl": 0.55157470703125,
      "learning_rate": 3.7288651837012745e-09,
      "loss": 0.0006,
      "reward": 1.377604205161333,
      "reward_std": 0.2037918628193438,
      "rewards/equation_reward_func": 0.42968751140870154,
      "rewards/format_reward_func": 0.9479166828095913,
      "step": 426
    },
    {
      "completion_length": 304.3593854904175,
      "epoch": 0.22826666666666667,
      "grad_norm": 0.14316318289656946,
      "kl": 0.23516845703125,
      "learning_rate": 3.134530561862081e-09,
      "loss": 0.0002,
      "reward": 1.4114583805203438,
      "reward_std": 0.2521082772873342,
      "rewards/equation_reward_func": 0.4687500149011612,
      "rewards/format_reward_func": 0.9427083544433117,
      "step": 428
    },
    {
      "completion_length": 299.0755262374878,
      "epoch": 0.22933333333333333,
      "grad_norm": 0.12545727015621191,
      "kl": 0.2213134765625,
      "learning_rate": 2.5914631388619103e-09,
      "loss": 0.0002,
      "reward": 1.4453125447034836,
      "reward_std": 0.18874157266691327,
      "rewards/equation_reward_func": 0.4869791781529784,
      "rewards/format_reward_func": 0.9583333507180214,
      "step": 430
    },
    {
      "completion_length": 265.3099060058594,
      "epoch": 0.2304,
      "grad_norm": 0.10719374815625937,
      "kl": 0.2373046875,
      "learning_rate": 2.0997756949353297e-09,
      "loss": 0.0002,
      "reward": 1.5390625298023224,
      "reward_std": 0.15120991226285696,
      "rewards/equation_reward_func": 0.5546875158324838,
      "rewards/format_reward_func": 0.9843750074505806,
      "step": 432
    },
    {
      "completion_length": 321.8854274749756,
      "epoch": 0.23146666666666665,
      "grad_norm": 0.1258623045036253,
      "kl": 0.20965576171875,
      "learning_rate": 1.6595703401020844e-09,
      "loss": 0.0002,
      "reward": 1.401041690260172,
      "reward_std": 0.23577453382313251,
      "rewards/equation_reward_func": 0.45312501303851604,
      "rewards/format_reward_func": 0.9479166902601719,
      "step": 434
    },
    {
      "completion_length": 322.6406316757202,
      "epoch": 0.23253333333333334,
      "grad_norm": 0.10043219855641766,
      "kl": 0.21881103515625,
      "learning_rate": 1.2709384929615596e-09,
      "loss": 0.0002,
      "reward": 1.3489583618938923,
      "reward_std": 0.24870790215209126,
      "rewards/equation_reward_func": 0.41145834466442466,
      "rewards/format_reward_func": 0.9375000186264515,
      "step": 436
    },
    {
      "completion_length": 335.68751335144043,
      "epoch": 0.2336,
      "grad_norm": 0.07244703059521732,
      "kl": 0.234619140625,
      "learning_rate": 9.339608617077165e-10,
      "loss": 0.0002,
      "reward": 1.380208358168602,
      "reward_std": 0.23140405910089612,
      "rewards/equation_reward_func": 0.4427083460614085,
      "rewards/format_reward_func": 0.9375000186264515,
      "step": 438
    },
    {
      "completion_length": 291.8854269981384,
      "epoch": 0.23466666666666666,
      "grad_norm": 0.08547530539534628,
      "kl": 0.23675537109375,
      "learning_rate": 6.487074273681114e-10,
      "loss": 0.0002,
      "reward": 1.4973958656191826,
      "reward_std": 0.15349498065188527,
      "rewards/equation_reward_func": 0.5234375093132257,
      "rewards/format_reward_func": 0.9739583469927311,
      "step": 440
    },
    {
      "completion_length": 312.1250047683716,
      "epoch": 0.23573333333333332,
      "grad_norm": 0.074850637657335,
      "kl": 0.22784423828125,
      "learning_rate": 4.152374292708538e-10,
      "loss": 0.0002,
      "reward": 1.3593750521540642,
      "reward_std": 0.22883992129936814,
      "rewards/equation_reward_func": 0.4192708432674408,
      "rewards/format_reward_func": 0.9401041902601719,
      "step": 442
    },
    {
      "completion_length": 294.8177156448364,
      "epoch": 0.2368,
      "grad_norm": 0.07224815686769165,
      "kl": 0.2313232421875,
      "learning_rate": 2.3359935274214204e-10,
      "loss": 0.0002,
      "reward": 1.4661458730697632,
      "reward_std": 0.1822908315807581,
      "rewards/equation_reward_func": 0.5026041779201478,
      "rewards/format_reward_func": 0.9635416828095913,
      "step": 444
    },
    {
      "completion_length": 323.0520935058594,
      "epoch": 0.23786666666666667,
      "grad_norm": 0.0603941305498588,
      "kl": 0.21807861328125,
      "learning_rate": 1.0383091903720665e-10,
      "loss": 0.0002,
      "reward": 1.4427083805203438,
      "reward_std": 0.21741202333942056,
      "rewards/equation_reward_func": 0.48177084885537624,
      "rewards/format_reward_func": 0.9609375149011612,
      "step": 446
    },
    {
      "completion_length": 299.1953172683716,
      "epoch": 0.23893333333333333,
      "grad_norm": 0.09701186524357072,
      "kl": 0.20562744140625,
      "learning_rate": 2.595907750671533e-11,
      "loss": 0.0002,
      "reward": 1.5156250298023224,
      "reward_std": 0.19597027264535427,
      "rewards/equation_reward_func": 0.5546875107102096,
      "rewards/format_reward_func": 0.9609375111758709,
      "step": 448
    },
    {
      "completion_length": 325.67709255218506,
      "epoch": 0.24,
      "grad_norm": 0.1517026640806461,
      "kl": 0.2586669921875,
      "learning_rate": 0.0,
      "loss": 0.0003,
      "reward": 1.3750000335276127,
      "reward_std": 0.18915646150708199,
      "rewards/equation_reward_func": 0.41666667629033327,
      "rewards/format_reward_func": 0.9583333432674408,
      "step": 450
    },
    {
      "epoch": 0.24,
      "step": 450,
      "total_flos": 0.0,
      "train_loss": 3.0447768125062188e-05,
      "train_runtime": 1961.2328,
      "train_samples_per_second": 5.507,
      "train_steps_per_second": 0.229
    }
  ],
  "logging_steps": 2,
  "max_steps": 450,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 50,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}